release: version 0.8.0 🚀

HISTORY.md

@@ -4,11 +4,73 @@ Changelog
 
 (unreleased)
 ------------
+- Feat: better dashboard reloading mechanism, refs NOISSUE. [Simon
+  Diesenreiter]
+- Feat: add explicit workflow steps and guardrail prompts, refs NOISSUE.
+  [Simon Diesenreiter]
+
+
+0.7.1 (2026-04-11)
+------------------
+
+Fix
+~~~
+- Add additional deletion confirmation, refs NOISSUE. [Simon
+  Diesenreiter]
+
+Other
+~~~~~
+
+
+0.7.0 (2026-04-10)
+------------------
+- Feat: gitea issue integration, refs NOISSUE. [Simon Diesenreiter]
+- Feat: better history data, refs NOISSUE. [Simon Diesenreiter]
+
+
+0.6.5 (2026-04-10)
+------------------
+
+Fix
+~~~
+- Better n8n workflow, refs NOISSUE. [Simon Diesenreiter]
+
+Other
+~~~~~
+
+
+0.6.4 (2026-04-10)
+------------------
+
+Fix
+~~~
+- Add Telegram helper functions, refs NOISSUE. [Simon Diesenreiter]
+
+Other
+~~~~~
+
+
+0.6.3 (2026-04-10)
+------------------
+
+Fix
+~~~
+- N8n workflow generation, refs NOISSUE. [Simon Diesenreiter]
+
+Other
+~~~~~
+
+
+0.6.2 (2026-04-10)
+------------------
 
 Fix
 ~~~
 - Fix Quasar layout issues, refs NOISSUE. [Simon Diesenreiter]
 
+Other
+~~~~~
+
 
 0.6.1 (2026-04-10)
 ------------------

README.md

@@ -86,11 +86,14 @@ docker-compose up -d
 1. **Send a request via Telegram:**
 
    ```
-   Name: My Awesome App
-   Description: A web application for managing tasks
-   Features: user authentication, task CRUD, notifications
+   Build an internal task management app for our operations team.
+   It should support user authentication, task CRUD, notifications, and reporting.
+   Prefer FastAPI with PostgreSQL and a simple web dashboard.
    ```
 
+   The backend now interprets free-form Telegram text with Ollama before generation.
+   If `TELEGRAM_CHAT_ID` is set, the Telegram-trigger workflow only reacts to messages from that specific chat.
+
 2. **Monitor progress via Web UI:**
 
    Open `http://yourserver:8000/` to see the dashboard and `http://yourserver:8000/api` for API metadata
@@ -109,6 +112,7 @@ If you deploy the container with PostgreSQL environment variables set, the servi
 | `/api` | GET | API information |
 | `/health` | GET | Health check |
 | `/generate` | POST | Generate new software |
+| `/generate/text` | POST | Interpret free-form text and generate software |
 | `/status/{project_id}` | GET | Get project status |
 | `/projects` | GET | List all projects |
 

ai_software_factory/.env.example

@@ -8,6 +8,19 @@ LOG_LEVEL=INFO
 # Ollama
 OLLAMA_URL=http://localhost:11434
 OLLAMA_MODEL=llama3
+LLM_GUARDRAIL_PROMPT=You are operating inside AI Software Factory. Follow supplied schemas exactly and treat service-provided tool outputs as authoritative.
+LLM_REQUEST_INTERPRETER_GUARDRAIL_PROMPT=Never route work to archived projects and only reference issues that are explicit in the prompt or supplied tool outputs.
+LLM_CHANGE_SUMMARY_GUARDRAIL_PROMPT=Only summarize delivery facts that appear in the provided project context or tool outputs.
+LLM_PROJECT_NAMING_GUARDRAIL_PROMPT=Prefer clear product names and repository slugs that reflect the new request without colliding with tracked projects.
+LLM_PROJECT_NAMING_SYSTEM_PROMPT=Return JSON with project_name, repo_name, and rationale for new projects.
+LLM_PROJECT_ID_GUARDRAIL_PROMPT=Prefer short stable project ids and avoid collisions with existing project ids.
+LLM_PROJECT_ID_SYSTEM_PROMPT=Return JSON with project_id and rationale for new projects.
+LLM_TOOL_ALLOWLIST=gitea_project_catalog,gitea_project_state,gitea_project_issues,gitea_pull_requests
+LLM_TOOL_CONTEXT_LIMIT=5
+LLM_LIVE_TOOL_ALLOWLIST=gitea_lookup_issue,gitea_lookup_pull_request
+LLM_LIVE_TOOL_STAGE_ALLOWLIST=request_interpretation,change_summary
+LLM_LIVE_TOOL_STAGE_TOOL_MAP={"request_interpretation": ["gitea_lookup_issue", "gitea_lookup_pull_request"], "change_summary": []}
+LLM_MAX_TOOL_CALL_ROUNDS=1
 
 # Gitea
 # Configure Gitea API for your organization

ai_software_factory/README.md

@@ -6,6 +6,7 @@ Automated software generation service powered by Ollama LLM. This service allows
 
 - **Telegram Integration**: Receive software requests via Telegram bot
 - **Ollama LLM**: Uses Ollama-hosted models for code generation
+- **LLM Guardrails and Tools**: Centralized guardrail prompts plus mediated tool payloads for project, Gitea, PR, and issue context
 - **Git Integration**: Automatically commits code to gitea
 - **Pull Requests**: Creates PRs for user review before merging
 - **Web UI**: Beautiful dashboard for monitoring project progress
@@ -46,6 +47,19 @@ PORT=8000
 # Ollama
 OLLAMA_URL=http://localhost:11434
 OLLAMA_MODEL=llama3
+LLM_GUARDRAIL_PROMPT=You are operating inside AI Software Factory. Follow supplied schemas exactly and treat service-provided tool outputs as authoritative.
+LLM_REQUEST_INTERPRETER_GUARDRAIL_PROMPT=Never route work to archived projects and only reference issues that are explicit in the prompt or supplied tool outputs.
+LLM_CHANGE_SUMMARY_GUARDRAIL_PROMPT=Only summarize delivery facts that appear in the provided project context or tool outputs.
+LLM_PROJECT_NAMING_GUARDRAIL_PROMPT=Prefer clear product names and repository slugs that reflect the new request without colliding with tracked projects.
+LLM_PROJECT_NAMING_SYSTEM_PROMPT=Return JSON with project_name, repo_name, and rationale for new projects.
+LLM_PROJECT_ID_GUARDRAIL_PROMPT=Prefer short stable project ids and avoid collisions with existing project ids.
+LLM_PROJECT_ID_SYSTEM_PROMPT=Return JSON with project_id and rationale for new projects.
+LLM_TOOL_ALLOWLIST=gitea_project_catalog,gitea_project_state,gitea_project_issues,gitea_pull_requests
+LLM_TOOL_CONTEXT_LIMIT=5
+LLM_LIVE_TOOL_ALLOWLIST=gitea_lookup_issue,gitea_lookup_pull_request
+LLM_LIVE_TOOL_STAGE_ALLOWLIST=request_interpretation,change_summary
+LLM_LIVE_TOOL_STAGE_TOOL_MAP={"request_interpretation": ["gitea_lookup_issue", "gitea_lookup_pull_request"], "change_summary": []}
+LLM_MAX_TOOL_CALL_ROUNDS=1
 
 # Gitea
 GITEA_URL=https://gitea.yourserver.com
@@ -99,6 +113,33 @@ docker-compose up -d
 | `/status/{project_id}` | GET | Get project status |
 | `/projects` | GET | List all projects |
 
+## LLM Guardrails and Tool Access
+
+External LLM calls are now routed through a centralized client that applies:
+
+- A global guardrail prompt for every outbound model request
+- Stage-specific guardrails for request interpretation and change summaries
+- Service-mediated tool outputs that expose tracked Gitea/project state without giving the model raw credentials
+
+Current mediated tools include:
+
+- `gitea_project_catalog`: active tracked projects and repository mappings
+- `gitea_project_state`: current repository, PR, and linked-issue state for the project in scope
+- `gitea_project_issues`: tracked open issues for the relevant repository
+- `gitea_pull_requests`: tracked pull requests for the relevant repository
+
+The service also supports a bounded live tool-call loop for selected lookups. When enabled, the model may request one live call such as `gitea_lookup_issue` or `gitea_lookup_pull_request`, the service executes it against Gitea, and the final model response is generated from the returned result. This remains mediated by the service, so the model never receives raw credentials.
+
+Live tool access is stage-aware. `LLM_LIVE_TOOL_ALLOWLIST` controls which live tools exist globally, while `LLM_LIVE_TOOL_STAGE_ALLOWLIST` controls which LLM stages may use them. If you need per-stage subsets, `LLM_LIVE_TOOL_STAGE_TOOL_MAP` accepts a JSON object mapping each stage to the exact tools it may use. For example, you can allow issue and PR lookups during `request_interpretation` while keeping `change_summary` fully read-only.
+
+When the interpreter decides a prompt starts a new project, the service can run a dedicated `project_naming` LLM stage before generation. `LLM_PROJECT_NAMING_SYSTEM_PROMPT` and `LLM_PROJECT_NAMING_GUARDRAIL_PROMPT` let you steer how project titles and repository slugs are chosen. The interpreter now checks tracked project repositories plus live Gitea repository names when available, so if the model suggests a colliding repo slug the service will automatically move to the next available slug.
+
+New project creation can also run a dedicated `project_id_naming` stage. `LLM_PROJECT_ID_SYSTEM_PROMPT` and `LLM_PROJECT_ID_GUARDRAIL_PROMPT` control how stable project ids are chosen, and the service will append deterministic numeric suffixes when an id is already taken instead of always falling back to a random UUID-based id.
+
+Runtime visibility for the active guardrails, mediated tools, live tools, and model configuration is available at `/llm/runtime` and in the dashboard System tab.
+
+These tool payloads are appended to the model prompt as authoritative JSON generated by the service, so the LLM can reason over live project and Gitea context while remaining constrained by the configured guardrails.
+
 ## Development
 
 ### Makefile Targets

ai_software_factory/VERSION

@@ -1 +1 @@
-0.6.2
+0.8.0

ai_software_factory/agents/change_summary.py

@@ -0,0 +1,125 @@
+"""Generate concise chat-friendly summaries of software generation results."""
+
+from __future__ import annotations
+
+try:
+    from ..config import settings
+    from .llm_service import LLMServiceClient
+except ImportError:
+    from config import settings
+    from agents.llm_service import LLMServiceClient
+
+
+class ChangeSummaryGenerator:
+    """Create a readable overview of generated changes for chat responses."""
+
+    def __init__(self, ollama_url: str | None = None, model: str | None = None):
+        self.ollama_url = (ollama_url or settings.ollama_url).rstrip('/')
+        self.model = model or settings.OLLAMA_MODEL
+        self.llm_client = LLMServiceClient(ollama_url=self.ollama_url, model=self.model)
+
+    async def summarize(self, context: dict) -> str:
+        """Summarize project changes with Ollama, or fall back to a deterministic overview."""
+        summary, _trace = await self.summarize_with_trace(context)
+        return summary
+
+    async def summarize_with_trace(self, context: dict) -> tuple[str, dict]:
+        """Summarize project changes with Ollama, or fall back to a deterministic overview."""
+        prompt = self._prompt(context)
+        system_prompt = (
+            'You write concise but informative mobile chat summaries of software delivery work. '
+            'Write 3 to 5 sentences. Mention the application goal, main delivered pieces, '
+            'technical direction, and what the user should expect next. Avoid markdown bullets.'
+        )
+        content, trace = await self.llm_client.chat_with_trace(
+            stage='change_summary',
+            system_prompt=system_prompt,
+            user_prompt=prompt,
+            tool_context_input={
+                'project_id': context.get('project_id'),
+                'project_name': context.get('name'),
+                'repository': context.get('repository'),
+                'repository_url': context.get('repository_url'),
+                'pull_request': context.get('pull_request'),
+                'pull_request_url': context.get('pull_request_url'),
+                'pull_request_state': context.get('pull_request_state'),
+                'related_issue': context.get('related_issue'),
+                'issues': [context.get('related_issue')] if context.get('related_issue') else [],
+            },
+        )
+        if content:
+            return content.strip(), trace
+
+        fallback = self._fallback(context)
+        return fallback, {
+            'stage': 'change_summary',
+            'provider': 'fallback',
+            'model': self.model,
+            'system_prompt': system_prompt,
+            'user_prompt': prompt,
+            'assistant_response': fallback,
+            'raw_response': {'fallback': 'deterministic', 'llm_trace': trace.get('raw_response') if isinstance(trace, dict) else None},
+            'guardrails': trace.get('guardrails') if isinstance(trace, dict) else [],
+            'tool_context': trace.get('tool_context') if isinstance(trace, dict) else [],
+            'fallback_used': True,
+        }
+
+    def _prompt(self, context: dict) -> str:
+        features = ', '.join(context.get('features') or []) or 'No explicit features recorded'
+        tech_stack = ', '.join(context.get('tech_stack') or []) or 'No explicit tech stack recorded'
+        changed_files = ', '.join(context.get('changed_files') or []) or 'No files recorded'
+        logs = ' | '.join((context.get('logs') or [])[:4]) or 'No log excerpts'
+        return (
+            f"Project name: {context.get('name', 'Unknown project')}\n"
+            f"Description: {context.get('description', '')}\n"
+            f"Features: {features}\n"
+            f"Tech stack: {tech_stack}\n"
+            f"Changed files: {changed_files}\n"
+            f"Repository: {context.get('repository_url') or 'No repository URL'}\n"
+            f"Pull request: {context.get('pull_request_url') or 'No pull request URL'}\n"
+            f"Pull request state: {context.get('pull_request_state') or 'No pull request state'}\n"
+            f"Status message: {context.get('message') or ''}\n"
+            f"Log excerpts: {logs}\n"
+            "Write a broad but phone-friendly summary of what was done."
+        )
+
+    def _fallback(self, context: dict) -> str:
+        name = context.get('name', 'The project')
+        description = context.get('description') or 'a software request'
+        changed_files = context.get('changed_files') or []
+        features = context.get('features') or []
+        tech_stack = context.get('tech_stack') or []
+        repo_url = context.get('repository_url')
+        repo_status = context.get('repository_status')
+        pr_url = context.get('pull_request_url')
+        pr_state = context.get('pull_request_state')
+
+        first_sentence = f"{name} was generated from your request for {description}."
+        feature_sentence = (
+            f"The delivery focused on {', '.join(features[:3])}."
+            if features else
+            "The delivery focused on turning the request into an initial runnable application skeleton."
+        )
+        tech_sentence = (
+            f"The generated implementation currently targets {', '.join(tech_stack[:3])}."
+            if tech_stack else
+            "The implementation was created with the current default stack configured for the factory."
+        )
+        file_sentence = (
+            f"Key artifacts were updated across {len(changed_files)} files, including {', '.join(changed_files[:3])}."
+            if changed_files else
+            "The service completed the generation flow, but no changed file list was returned."
+        )
+        if repo_url:
+            repo_sentence = f"The resulting project is tracked at {repo_url}."
+        elif repo_status in {'pending', 'skipped', 'error'}:
+            repo_sentence = "Repository provisioning was not confirmed, so review the Gitea status in the dashboard before assuming a remote repo exists."
+        else:
+            repo_sentence = "The project is ready for further review in the dashboard."
+        if pr_url and pr_state == 'open':
+            pr_sentence = f"An open pull request is ready for review at {pr_url}, and later prompts will continue updating that same PR until it is merged."
+        elif pr_url:
+            pr_sentence = f"The latest pull request is available at {pr_url}."
+        else:
+            pr_sentence = "No pull request link was recorded for this delivery."
+        return ' '.join([first_sentence, feature_sentence, tech_sentence, file_sentence, repo_sentence, pr_sentence])

ai_software_factory/agents/git_manager.py

@@ -1,7 +1,9 @@
 """Git manager for project operations."""
 
 import os
+import shutil
 import subprocess
+import tempfile
 from pathlib import Path
 from typing import Optional
 
@@ -14,53 +16,140 @@ except ImportError:
 class GitManager:
     """Manages git operations for the project."""
     
-    def __init__(self, project_id: str):
+    def __init__(self, project_id: str, project_dir: str | None = None):
         if not project_id:
             raise ValueError("project_id cannot be empty or None")
         self.project_id = project_id
-        project_path = Path(project_id)
-        if project_path.is_absolute() or len(project_path.parts) > 1:
-            resolved = project_path.expanduser().resolve()
+        if project_dir:
+            resolved = Path(project_dir).expanduser().resolve()
         else:
-            base_root = settings.projects_root
-            if base_root.name != "test-project":
-                base_root = base_root / "test-project"
-            resolved = (base_root / project_id).resolve()
+            project_path = Path(project_id)
+            if project_path.is_absolute() or len(project_path.parts) > 1:
+                resolved = project_path.expanduser().resolve()
+            else:
+                base_root = settings.projects_root
+                if base_root.name != "test-project":
+                    base_root = base_root / "test-project"
+                resolved = (base_root / project_id).resolve()
         self.project_dir = str(resolved)
+
+    def is_git_available(self) -> bool:
+        """Return whether the git executable is available in the current environment."""
+        return shutil.which('git') is not None
+
+    def _ensure_git_available(self) -> None:
+        """Raise a clear error when git is not installed in the runtime environment."""
+        if not self.is_git_available():
+            raise RuntimeError('git executable is not available in PATH')
+
+    def _run(self, args: list[str], env: dict | None = None, check: bool = True) -> subprocess.CompletedProcess:
+        """Run a git command in the project directory."""
+        self._ensure_git_available()
+        return subprocess.run(
+            args,
+            check=check,
+            capture_output=True,
+            text=True,
+            cwd=self.project_dir,
+            env=env,
+        )
+
+    def has_repo(self) -> bool:
+        """Return whether the project directory already contains a git repository."""
+        return Path(self.project_dir, '.git').exists()
     
     def init_repo(self):
         """Initialize git repository."""
         os.makedirs(self.project_dir, exist_ok=True)
-        os.chdir(self.project_dir)
-        subprocess.run(["git", "init"], check=True, capture_output=True)
+        self._run(["git", "init", "-b", "main"])
+        self._run(["git", "config", "user.name", "AI Software Factory"])
+        self._run(["git", "config", "user.email", "factory@local.invalid"])
     
     def add_files(self, paths: list[str]):
         """Add files to git staging."""
-        subprocess.run(["git", "add"] + paths, check=True, capture_output=True)
+        self._run(["git", "add"] + paths)
+
+    def checkout_branch(self, branch_name: str, create: bool = False, start_point: str | None = None) -> None:
+        """Switch to a branch, optionally creating it from a start point."""
+        if create:
+            args = ["git", "checkout", "-B", branch_name]
+            if start_point:
+                args.append(start_point)
+            self._run(args)
+            return
+        self._run(["git", "checkout", branch_name])
+
+    def branch_exists(self, branch_name: str) -> bool:
+        """Return whether a local branch exists."""
+        result = self._run(["git", "show-ref", "--verify", f"refs/heads/{branch_name}"], check=False)
+        return result.returncode == 0
     
-    def commit(self, message: str):
+    def commit(self, message: str) -> str:
         """Create a git commit."""
-        subprocess.run(
-            ["git", "commit", "-m", message],
-            check=True,
-            capture_output=True
-        )
+        self._run(["git", "commit", "-m", message])
+        return self.current_head()
+
+    def create_empty_commit(self, message: str) -> str:
+        """Create an empty commit."""
+        self._run(["git", "commit", "--allow-empty", "-m", message])
+        return self.current_head()
     
     def push(self, remote: str = "origin", branch: str = "main"):
         """Push changes to remote."""
-        subprocess.run(
-            ["git", "push", "-u", remote, branch],
-            check=True,
-            capture_output=True
-        )
+        self._run(["git", "push", "-u", remote, branch])
+
+    def ensure_remote(self, remote: str, url: str) -> None:
+        """Create or update a remote URL."""
+        result = self._run(["git", "remote", "get-url", remote], check=False)
+        if result.returncode == 0:
+            self._run(["git", "remote", "set-url", remote, url])
+        else:
+            self._run(["git", "remote", "add", remote, url])
+
+    def push_with_credentials(
+        self,
+        remote_url: str,
+        username: str,
+        password: str,
+        remote: str = "origin",
+        branch: str = "main",
+    ) -> None:
+        """Push to a remote over HTTPS using an askpass helper."""
+        os.makedirs(self.project_dir, exist_ok=True)
+        self.ensure_remote(remote, remote_url)
+        helper_contents = "#!/bin/sh\ncase \"$1\" in\n  *Username*) printf '%s\\n' \"$GIT_ASKPASS_USERNAME\" ;;\n  *) printf '%s\\n' \"$GIT_ASKPASS_PASSWORD\" ;;\nesac\n"
+        helper_path: str | None = None
+        try:
+            with tempfile.NamedTemporaryFile('w', delete=False, dir=self.project_dir, prefix='git-askpass-', suffix='.sh') as helper_file:
+                helper_file.write(helper_contents)
+                helper_path = helper_file.name
+            os.chmod(helper_path, 0o700)
+            env = os.environ.copy()
+            env.update(
+                {
+                    "GIT_TERMINAL_PROMPT": "0",
+                    "GIT_ASKPASS": helper_path,
+                    "GIT_ASKPASS_USERNAME": username,
+                    "GIT_ASKPASS_PASSWORD": password,
+                }
+            )
+            self._run(["git", "push", "-u", remote, branch], env=env)
+        finally:
+            if helper_path:
+                Path(helper_path).unlink(missing_ok=True)
     
     def create_branch(self, branch_name: str):
         """Create and switch to a new branch."""
-        subprocess.run(
-            ["git", "checkout", "-b", branch_name],
-            check=True,
-            capture_output=True
-        )
+        self._run(["git", "checkout", "-b", branch_name])
+
+    def revert_commit(self, commit_hash: str, no_edit: bool = True) -> str:
+        """Revert a commit and return the new HEAD."""
+        args = ["git", "revert"]
+        if no_edit:
+            args.append("--no-edit")
+        args.append(commit_hash)
+        self._run(args)
+        return self.current_head()
     
     def create_pr(
         self,
@@ -84,6 +173,18 @@ class GitManager:
         result = subprocess.run(
             ["git", "status", "--porcelain"],
             capture_output=True,
-            text=True
+            text=True,
+            cwd=self.project_dir,
         )
         return result.stdout.strip()
+
+    def current_head(self) -> str:
+        """Return the current commit hash."""
+        return self._run(["git", "rev-parse", "HEAD"]).stdout.strip()
+
+    def current_head_or_none(self) -> str | None:
+        """Return the current commit hash when the repository already has commits."""
+        result = self._run(["git", "rev-parse", "HEAD"], check=False)
+        if result.returncode != 0:
+            return None
+        return result.stdout.strip() or None

ai_software_factory/agents/gitea.py

@@ -1,6 +1,9 @@
 """Gitea API integration for repository and pull request operations."""
 
 import os
+import urllib.error
+import urllib.request
+import json
 
 
 class GiteaAPI:
@@ -41,6 +44,38 @@ class GiteaAPI:
         """Build a Gitea API URL from a relative path."""
         return f"{self.base_url}/api/v1/{path.lstrip('/')}"
 
+    def build_repo_git_url(self, owner: str | None = None, repo: str | None = None) -> str | None:
+        """Build the clone URL for a repository."""
+        _owner = owner or self.owner
+        _repo = repo or self.repo
+        if not _owner or not _repo:
+            return None
+        return f"{self.base_url}/{_owner}/{_repo}.git"
+
+    def build_commit_url(self, commit_hash: str, owner: str | None = None, repo: str | None = None) -> str | None:
+        """Build a browser URL for a commit."""
+        _owner = owner or self.owner
+        _repo = repo or self.repo
+        if not _owner or not _repo or not commit_hash:
+            return None
+        return f"{self.base_url}/{_owner}/{_repo}/commit/{commit_hash}"
+
+    def build_compare_url(self, base_ref: str, head_ref: str, owner: str | None = None, repo: str | None = None) -> str | None:
+        """Build a browser URL for a compare view."""
+        _owner = owner or self.owner
+        _repo = repo or self.repo
+        if not _owner or not _repo or not base_ref or not head_ref:
+            return None
+        return f"{self.base_url}/{_owner}/{_repo}/compare/{base_ref}...{head_ref}"
+
+    def build_pull_request_url(self, pr_number: int, owner: str | None = None, repo: str | None = None) -> str | None:
+        """Build a browser URL for a pull request."""
+        _owner = owner or self.owner
+        _repo = repo or self.repo
+        if not _owner or not _repo or not pr_number:
+            return None
+        return f"{self.base_url}/{_owner}/{_repo}/pulls/{pr_number}"
+
     async def _request(self, method: str, path: str, payload: dict | None = None) -> dict:
         """Perform a Gitea API request and normalize the response."""
         try:
@@ -59,6 +94,30 @@ class GiteaAPI:
         except Exception as e:
             return {"error": str(e)}
 
+    def _request_sync(self, method: str, path: str, payload: dict | None = None) -> dict:
+        """Perform a synchronous Gitea API request."""
+        request = urllib.request.Request(
+            self._api_url(path),
+            headers=self.get_auth_headers(),
+            method=method.upper(),
+        )
+        data = None
+        if payload is not None:
+            data = json.dumps(payload).encode('utf-8')
+            request.data = data
+        try:
+            with urllib.request.urlopen(request) as response:
+                body = response.read().decode('utf-8')
+                return json.loads(body) if body else {}
+        except urllib.error.HTTPError as exc:
+            try:
+                body = exc.read().decode('utf-8')
+            except Exception:
+                body = str(exc)
+            return {'error': body, 'status_code': exc.code}
+        except Exception as exc:
+            return {'error': str(exc)}
+
     def build_project_repo_name(self, project_id: str, project_name: str | None = None) -> str:
         """Build a repository name for a generated project."""
         preferred = (project_name or project_id or "project").strip().lower().replace(" ", "-")
@@ -97,6 +156,32 @@ class GiteaAPI:
             result.setdefault("status", "created")
         return result
 
+    async def delete_repo(self, owner: str | None = None, repo: str | None = None) -> dict:
+        """Delete a repository from the configured organization/user."""
+        _owner = owner or self.owner
+        _repo = repo or self.repo
+        if not _owner or not _repo:
+            return {'error': 'Owner and repository name are required'}
+        result = await self._request('DELETE', f'repos/{_owner}/{_repo}')
+        if not result.get('error'):
+            result.setdefault('status', 'deleted')
+        return result
+
+    def delete_repo_sync(self, owner: str | None = None, repo: str | None = None) -> dict:
+        """Synchronously delete a repository from the configured organization/user."""
+        _owner = owner or self.owner
+        _repo = repo or self.repo
+        if not _owner or not _repo:
+            return {'error': 'Owner and repository name are required'}
+        result = self._request_sync('DELETE', f'repos/{_owner}/{_repo}')
+        if not result.get('error'):
+            result.setdefault('status', 'deleted')
+        return result
+
+    async def get_current_user(self) -> dict:
+        """Get the user associated with the configured token."""
+        return await self._request("GET", "user")
+
     async def create_branch(self, branch: str, base: str = "main", owner: str | None = None, repo: str | None = None):
         """Create a new branch."""
         _owner = owner or self.owner
@@ -127,6 +212,144 @@ class GiteaAPI:
         }
         return await self._request("POST", f"repos/{_owner}/{_repo}/pulls", payload)
 
+    async def list_pull_requests(
+        self,
+        owner: str | None = None,
+        repo: str | None = None,
+        state: str = 'open',
+    ) -> dict | list:
+        """List pull requests for a repository."""
+        _owner = owner or self.owner
+        _repo = repo or self.repo
+        return await self._request("GET", f"repos/{_owner}/{_repo}/pulls?state={state}")
+
+    def list_pull_requests_sync(
+        self,
+        owner: str | None = None,
+        repo: str | None = None,
+        state: str = 'open',
+    ) -> dict | list:
+        """Synchronously list pull requests for a repository."""
+        _owner = owner or self.owner
+        _repo = repo or self.repo
+        return self._request_sync("GET", f"repos/{_owner}/{_repo}/pulls?state={state}")
+
+    async def list_repositories(self, owner: str | None = None) -> dict | list:
+        """List repositories within the configured organization."""
+        _owner = owner or self.owner
+        return await self._request("GET", f"orgs/{_owner}/repos")
+
+    def list_repositories_sync(self, owner: str | None = None) -> dict | list:
+        """Synchronously list repositories within the configured organization."""
+        _owner = owner or self.owner
+        return self._request_sync("GET", f"orgs/{_owner}/repos")
+
+    async def list_branches(self, owner: str | None = None, repo: str | None = None) -> dict | list:
+        """List repository branches."""
+        _owner = owner or self.owner
+        _repo = repo or self.repo
+        return await self._request("GET", f"repos/{_owner}/{_repo}/branches")
+
+    def list_branches_sync(self, owner: str | None = None, repo: str | None = None) -> dict | list:
+        """Synchronously list repository branches."""
+        _owner = owner or self.owner
+        _repo = repo or self.repo
+        return self._request_sync("GET", f"repos/{_owner}/{_repo}/branches")
+
+    async def list_issues(
+        self,
+        owner: str | None = None,
+        repo: str | None = None,
+        state: str = 'open',
+    ) -> dict | list:
+        """List repository issues, excluding pull requests at the consumer layer."""
+        _owner = owner or self.owner
+        _repo = repo or self.repo
+        return await self._request("GET", f"repos/{_owner}/{_repo}/issues?state={state}")
+
+    def list_issues_sync(
+        self,
+        owner: str | None = None,
+        repo: str | None = None,
+        state: str = 'open',
+    ) -> dict | list:
+        """Synchronously list repository issues."""
+        _owner = owner or self.owner
+        _repo = repo or self.repo
+        return self._request_sync("GET", f"repos/{_owner}/{_repo}/issues?state={state}")
+
+    async def get_issue(self, issue_number: int, owner: str | None = None, repo: str | None = None) -> dict:
+        """Return one repository issue by number."""
+        _owner = owner or self.owner
+        _repo = repo or self.repo
+        return await self._request("GET", f"repos/{_owner}/{_repo}/issues/{issue_number}")
+
+    def get_issue_sync(self, issue_number: int, owner: str | None = None, repo: str | None = None) -> dict:
+        """Synchronously return one repository issue by number."""
+        _owner = owner or self.owner
+        _repo = repo or self.repo
+        return self._request_sync("GET", f"repos/{_owner}/{_repo}/issues/{issue_number}")
+
+    async def list_repo_commits(
+        self,
+        owner: str | None = None,
+        repo: str | None = None,
+        limit: int = 25,
+        branch: str | None = None,
+    ) -> dict | list:
+        """List recent commits for a repository."""
+        _owner = owner or self.owner
+        _repo = repo or self.repo
+        branch_query = f"&sha={branch}" if branch else ""
+        return await self._request("GET", f"repos/{_owner}/{_repo}/commits?limit={limit}{branch_query}")
+
+    def list_repo_commits_sync(
+        self,
+        owner: str | None = None,
+        repo: str | None = None,
+        limit: int = 25,
+        branch: str | None = None,
+    ) -> dict | list:
+        """Synchronously list recent commits for a repository."""
+        _owner = owner or self.owner
+        _repo = repo or self.repo
+        branch_query = f"&sha={branch}" if branch else ""
+        return self._request_sync("GET", f"repos/{_owner}/{_repo}/commits?limit={limit}{branch_query}")
+
+    async def get_commit(
+        self,
+        commit_hash: str,
+        owner: str | None = None,
+        repo: str | None = None,
+    ) -> dict:
+        """Return one commit by hash."""
+        _owner = owner or self.owner
+        _repo = repo or self.repo
+        return await self._request("GET", f"repos/{_owner}/{_repo}/git/commits/{commit_hash}")
+
+    def get_commit_sync(
+        self,
+        commit_hash: str,
+        owner: str | None = None,
+        repo: str | None = None,
+    ) -> dict:
+        """Synchronously return one commit by hash."""
+        _owner = owner or self.owner
+        _repo = repo or self.repo
+        return self._request_sync("GET", f"repos/{_owner}/{_repo}/git/commits/{commit_hash}")
+
+    async def get_pull_request(self, pr_number: int, owner: str | None = None, repo: str | None = None) -> dict:
+        """Return one pull request by number."""
+        _owner = owner or self.owner
+        _repo = repo or self.repo
+        return await self._request("GET", f"repos/{_owner}/{_repo}/pulls/{pr_number}")
+
+    def get_pull_request_sync(self, pr_number: int, owner: str | None = None, repo: str | None = None) -> dict:
+        """Synchronously return one pull request by number."""
+        _owner = owner or self.owner
+        _repo = repo or self.repo
+        return self._request_sync("GET", f"repos/{_owner}/{_repo}/pulls/{pr_number}")
+
     async def push_commit(
         self,
         branch: str,

ai_software_factory/agents/llm_service.py

@@ -0,0 +1,394 @@
+"""Centralized LLM client with guardrails and mediated tool context."""
+
+from __future__ import annotations
+
+import json
+
+try:
+    from .gitea import GiteaAPI
+except ImportError:
+    from gitea import GiteaAPI
+
+try:
+    from ..config import settings
+except ImportError:
+    from config import settings
+
+
+class LLMToolbox:
+    """Build named tool payloads that can be shared with external LLM providers."""
+
+    SUPPORTED_LIVE_TOOL_STAGES = ('request_interpretation', 'change_summary', 'generation_plan', 'project_naming', 'project_id_naming')
+
+    def build_tool_context(self, stage: str, context: dict | None = None) -> list[dict]:
+        """Return the mediated tool payloads allowed for this LLM request."""
+        context = context or {}
+        allowed = set(settings.llm_tool_allowlist)
+        limit = settings.llm_tool_context_limit
+        tool_context: list[dict] = []
+
+        if 'gitea_project_catalog' in allowed:
+            projects = context.get('projects') or []
+            if projects:
+                tool_context.append(
+                    {
+                        'name': 'gitea_project_catalog',
+                        'description': 'Tracked active projects and their repository mappings inside the factory.',
+                        'payload': projects[:limit],
+                    }
+                )
+
+        if 'gitea_project_state' in allowed:
+            state_payload = {
+                'project_id': context.get('project_id'),
+                'project_name': context.get('project_name') or context.get('name'),
+                'repository': context.get('repository'),
+                'repository_url': context.get('repository_url'),
+                'pull_request': context.get('pull_request'),
+                'pull_request_url': context.get('pull_request_url'),
+                'pull_request_state': context.get('pull_request_state'),
+                'related_issue': context.get('related_issue'),
+            }
+            if any(value for value in state_payload.values()):
+                tool_context.append(
+                    {
+                        'name': 'gitea_project_state',
+                        'description': 'Current repository and pull-request state for the project being discussed.',
+                        'payload': state_payload,
+                    }
+                )
+
+        if 'gitea_project_issues' in allowed:
+            issues = context.get('open_issues') or context.get('issues') or []
+            if issues:
+                tool_context.append(
+                    {
+                        'name': 'gitea_project_issues',
+                        'description': 'Open tracked Gitea issues for the relevant project repository.',
+                        'payload': issues[:limit],
+                    }
+                )
+
+        if 'gitea_pull_requests' in allowed:
+            pull_requests = context.get('pull_requests') or []
+            if pull_requests:
+                tool_context.append(
+                    {
+                        'name': 'gitea_pull_requests',
+                        'description': 'Tracked pull requests associated with the relevant project repository.',
+                        'payload': pull_requests[:limit],
+                    }
+                )
+
+        return tool_context
+
+    def build_live_tool_specs(self, stage: str, context: dict | None = None) -> list[dict]:
+        """Return live tool-call specs that the model may request explicitly."""
+        _context = context or {}
+        specs = []
+        allowed = set(settings.llm_live_tools_for_stage(stage))
+        if 'gitea_lookup_issue' in allowed:
+            specs.append(
+                {
+                    'name': 'gitea_lookup_issue',
+                    'description': 'Fetch one live Gitea issue by issue number for a tracked repository.',
+                    'arguments': {
+                        'project_id': 'optional tracked project id',
+                        'owner': 'optional repository owner override',
+                        'repo': 'optional repository name override',
+                        'issue_number': 'required integer issue number',
+                    },
+                }
+            )
+        if 'gitea_lookup_pull_request' in allowed:
+            specs.append(
+                {
+                    'name': 'gitea_lookup_pull_request',
+                    'description': 'Fetch one live Gitea pull request by PR number for a tracked repository.',
+                    'arguments': {
+                        'project_id': 'optional tracked project id',
+                        'owner': 'optional repository owner override',
+                        'repo': 'optional repository name override',
+                        'pr_number': 'required integer pull request number',
+                    },
+                }
+            )
+        return specs
+
+
+class LLMLiveToolExecutor:
+    """Resolve bounded live tool requests on behalf of the model."""
+
+    def __init__(self):
+        self.gitea_api = None
+        if settings.gitea_url and settings.gitea_token:
+            self.gitea_api = GiteaAPI(
+                token=settings.GITEA_TOKEN,
+                base_url=settings.GITEA_URL,
+                owner=settings.GITEA_OWNER,
+                repo=settings.GITEA_REPO or '',
+            )
+
+    async def execute(self, tool_name: str, arguments: dict, context: dict | None = None) -> dict:
+        """Execute one live tool request and normalize the result."""
+        if tool_name not in set(settings.llm_live_tool_allowlist):
+            return {'error': f'Tool {tool_name} is not enabled'}
+        if self.gitea_api is None:
+            return {'error': 'Gitea live tool execution is not configured'}
+        resolved = self._resolve_repository(arguments=arguments, context=context or {})
+        if resolved.get('error'):
+            return resolved
+        owner = resolved['owner']
+        repo = resolved['repo']
+
+        if tool_name == 'gitea_lookup_issue':
+            issue_number = arguments.get('issue_number')
+            if issue_number is None:
+                return {'error': 'issue_number is required'}
+            return await self.gitea_api.get_issue(issue_number=int(issue_number), owner=owner, repo=repo)
+
+        if tool_name == 'gitea_lookup_pull_request':
+            pr_number = arguments.get('pr_number')
+            if pr_number is None:
+                return {'error': 'pr_number is required'}
+            return await self.gitea_api.get_pull_request(pr_number=int(pr_number), owner=owner, repo=repo)
+
+        return {'error': f'Unsupported tool {tool_name}'}
+
+    def _resolve_repository(self, arguments: dict, context: dict) -> dict:
+        """Resolve repository owner/name from explicit args or tracked project context."""
+        owner = arguments.get('owner')
+        repo = arguments.get('repo')
+        if owner and repo:
+            return {'owner': owner, 'repo': repo}
+        project_id = arguments.get('project_id')
+        if project_id:
+            for project in context.get('projects', []):
+                if project.get('project_id') == project_id:
+                    repository = project.get('repository') or {}
+                    if repository.get('owner') and repository.get('name'):
+                        return {'owner': repository['owner'], 'repo': repository['name']}
+            state = context.get('repository') or {}
+            if context.get('project_id') == project_id and state.get('owner') and state.get('name'):
+                return {'owner': state['owner'], 'repo': state['name']}
+        repository = context.get('repository') or {}
+        if repository.get('owner') and repository.get('name'):
+            return {'owner': repository['owner'], 'repo': repository['name']}
+        return {'error': 'Could not resolve repository for tool request'}
+
+
+class LLMServiceClient:
+    """Call the configured LLM provider with consistent guardrails and tool payloads."""
+
+    def __init__(self, ollama_url: str | None = None, model: str | None = None):
+        self.ollama_url = (ollama_url or settings.ollama_url).rstrip('/')
+        self.model = model or settings.OLLAMA_MODEL
+        self.toolbox = LLMToolbox()
+        self.live_tool_executor = LLMLiveToolExecutor()
+
+    async def chat_with_trace(
+        self,
+        *,
+        stage: str,
+        system_prompt: str,
+        user_prompt: str,
+        tool_context_input: dict | None = None,
+        expect_json: bool = False,
+    ) -> tuple[str | None, dict]:
+        """Invoke the configured LLM and return both content and a structured trace."""
+        effective_system_prompt = self._compose_system_prompt(stage, system_prompt)
+        tool_context = self.toolbox.build_tool_context(stage=stage, context=tool_context_input)
+        live_tool_specs = self.toolbox.build_live_tool_specs(stage=stage, context=tool_context_input)
+        effective_user_prompt = self._compose_user_prompt(user_prompt, tool_context, live_tool_specs)
+        raw_responses: list[dict] = []
+        executed_tool_calls: list[dict] = []
+        current_user_prompt = effective_user_prompt
+        max_rounds = settings.llm_max_tool_call_rounds
+
+        for round_index in range(max_rounds + 1):
+            content, payload, error = await self._send_chat_request(
+                system_prompt=effective_system_prompt,
+                user_prompt=current_user_prompt,
+                expect_json=expect_json,
+            )
+            raw_responses.append(payload)
+            if content:
+                tool_request = self._extract_tool_request(content)
+                if tool_request and round_index < max_rounds:
+                    tool_name = tool_request.get('name')
+                    tool_arguments = tool_request.get('arguments') or {}
+                    tool_result = await self.live_tool_executor.execute(tool_name, tool_arguments, tool_context_input)
+                    executed_tool_calls.append(
+                        {
+                            'name': tool_name,
+                            'arguments': tool_arguments,
+                            'result': tool_result,
+                        }
+                    )
+                    current_user_prompt = self._compose_follow_up_prompt(user_prompt, tool_context, live_tool_specs, executed_tool_calls)
+                    continue
+                return content, {
+                    'stage': stage,
+                    'provider': 'ollama',
+                    'model': self.model,
+                    'system_prompt': effective_system_prompt,
+                    'user_prompt': current_user_prompt,
+                    'assistant_response': content,
+                    'raw_response': {
+                        'provider_response': raw_responses[-1],
+                        'provider_responses': raw_responses,
+                        'tool_context': tool_context,
+                        'live_tool_specs': live_tool_specs,
+                        'executed_tool_calls': executed_tool_calls,
+                    },
+                    'raw_responses': raw_responses,
+                    'fallback_used': False,
+                    'guardrails': self._guardrail_sections(stage),
+                    'tool_context': tool_context,
+                    'live_tool_specs': live_tool_specs,
+                    'executed_tool_calls': executed_tool_calls,
+                }
+            if error:
+                break
+
+        return None, {
+            'stage': stage,
+            'provider': 'ollama',
+            'model': self.model,
+            'system_prompt': effective_system_prompt,
+            'user_prompt': current_user_prompt,
+            'assistant_response': '',
+            'raw_response': {
+                'provider_response': raw_responses[-1] if raw_responses else {'error': 'No response'},
+                'provider_responses': raw_responses,
+                'tool_context': tool_context,
+                'live_tool_specs': live_tool_specs,
+                'executed_tool_calls': executed_tool_calls,
+            },
+            'raw_responses': raw_responses,
+            'fallback_used': True,
+            'guardrails': self._guardrail_sections(stage),
+            'tool_context': tool_context,
+            'live_tool_specs': live_tool_specs,
+            'executed_tool_calls': executed_tool_calls,
+        }
+
+    async def _send_chat_request(self, *, system_prompt: str, user_prompt: str, expect_json: bool) -> tuple[str | None, dict, str | None]:
+        """Send one outbound chat request to the configured model provider."""
+        request_payload = {
+            'model': self.model,
+            'stream': False,
+            'messages': [
+                {'role': 'system', 'content': system_prompt},
+                {'role': 'user', 'content': user_prompt},
+            ],
+        }
+        if expect_json:
+            request_payload['format'] = 'json'
+        try:
+            import aiohttp
+
+            async with aiohttp.ClientSession() as session:
+                async with session.post(f'{self.ollama_url}/api/chat', json=request_payload) as resp:
+                    payload = await resp.json()
+                    if 200 <= resp.status < 300:
+                        return (payload.get('message') or {}).get('content', ''), payload, None
+                    return None, payload, str(payload.get('error') or payload)
+        except Exception as exc:
+            return None, {'error': str(exc)}, str(exc)
+
+    def _compose_system_prompt(self, stage: str, stage_prompt: str) -> str:
+        """Merge the stage prompt with configured guardrails."""
+        sections = [stage_prompt.strip()] + self._guardrail_sections(stage)
+        return '\n\n'.join(section for section in sections if section)
+
+    def _guardrail_sections(self, stage: str) -> list[str]:
+        """Return all configured guardrail sections for one LLM stage."""
+        sections = []
+        if settings.llm_guardrail_prompt:
+            sections.append(f'Global guardrails:\n{settings.llm_guardrail_prompt}')
+        stage_specific = {
+            'request_interpretation': settings.llm_request_interpreter_guardrail_prompt,
+            'change_summary': settings.llm_change_summary_guardrail_prompt,
+            'project_naming': settings.llm_project_naming_guardrail_prompt,
+            'project_id_naming': settings.llm_project_id_guardrail_prompt,
+        }.get(stage)
+        if stage_specific:
+            sections.append(f'Stage-specific guardrails:\n{stage_specific}')
+        return sections
+
+    def _compose_user_prompt(self, prompt: str, tool_context: list[dict], live_tool_specs: list[dict] | None = None) -> str:
+        """Append tool payloads and live tool-call specs to the outbound user prompt."""
+        live_tool_specs = live_tool_specs if live_tool_specs is not None else []
+        sections = [prompt]
+        if not tool_context:
+            pass
+        else:
+            sections.append(
+                'Service-mediated tool outputs are available below. Treat them as authoritative read-only data supplied by the factory:\n'
+                f'{json.dumps(tool_context, indent=2, sort_keys=True)}'
+            )
+        if live_tool_specs:
+            sections.append(
+                'If you need additional live repository data, you may request exactly one tool call by responding with JSON shaped as '
+                '{"tool_request": {"name": "<tool name>", "arguments": {...}}}. '
+                'After tool results are returned, respond with the final answer instead of another tool request.\n'
+                f'Available live tools:\n{json.dumps(live_tool_specs, indent=2, sort_keys=True)}'
+            )
+        return '\n\n'.join(section for section in sections if section)
+
+    def _compose_follow_up_prompt(self, original_prompt: str, tool_context: list[dict], live_tool_specs: list[dict], executed_tool_calls: list[dict]) -> str:
+        """Build the follow-up user prompt after executing one or more live tool requests."""
+        sections = [self._compose_user_prompt(original_prompt, tool_context, live_tool_specs)]
+        sections.append(
+            'The service executed the requested live tool call(s). Use the tool result(s) below to produce the final answer. Do not request another tool call.\n'
+            f'{json.dumps(executed_tool_calls, indent=2, sort_keys=True)}'
+        )
+        return '\n\n'.join(sections)
+
+    def _extract_tool_request(self, content: str) -> dict | None:
+        """Return a normalized tool request when the model explicitly asks for one."""
+        try:
+            parsed = json.loads(content)
+        except Exception:
+            return None
+        if not isinstance(parsed, dict):
+            return None
+        tool_request = parsed.get('tool_request')
+        if not isinstance(tool_request, dict) or not tool_request.get('name'):
+            return None
+        return {
+            'name': str(tool_request.get('name')).strip(),
+            'arguments': tool_request.get('arguments') or {},
+        }
+
+    def get_runtime_configuration(self) -> dict:
+        """Return the active LLM runtime config, guardrails, and tool exposure."""
+        live_tool_stages = {
+            stage: settings.llm_live_tools_for_stage(stage)
+            for stage in self.toolbox.SUPPORTED_LIVE_TOOL_STAGES
+        }
+        return {
+            'provider': 'ollama',
+            'ollama_url': self.ollama_url,
+            'model': self.model,
+            'guardrails': {
+                'global': settings.llm_guardrail_prompt,
+                'request_interpretation': settings.llm_request_interpreter_guardrail_prompt,
+                'change_summary': settings.llm_change_summary_guardrail_prompt,
+                'project_naming': settings.llm_project_naming_guardrail_prompt,
+                'project_id_naming': settings.llm_project_id_guardrail_prompt,
+            },
+            'system_prompts': {
+                'project_naming': settings.llm_project_naming_system_prompt,
+                'project_id_naming': settings.llm_project_id_system_prompt,
+            },
+            'mediated_tools': settings.llm_tool_allowlist,
+            'live_tools': settings.llm_live_tool_allowlist,
+            'live_tool_stage_allowlist': settings.llm_live_tool_stage_allowlist,
+            'live_tool_stage_tool_map': settings.llm_live_tool_stage_tool_map,
+            'live_tools_by_stage': live_tool_stages,
+            'tool_context_limit': settings.llm_tool_context_limit,
+            'max_tool_call_rounds': settings.llm_max_tool_call_rounds,
+            'gitea_live_tools_configured': bool(settings.gitea_url and settings.gitea_token),
+        }

ai_software_factory/agents/n8n_setup.py

@@ -220,11 +220,31 @@ class N8NSetupAgent:
     
     async def create_workflow(self, workflow_json: dict) -> dict:
         """Create or update a workflow."""
-        return await self._request("POST", "workflows", json=workflow_json)
+        return await self._request("POST", "workflows", json=self._workflow_payload(workflow_json))
+
+    def _workflow_payload(self, workflow_json: dict) -> dict:
+        """Return a workflow payload without server-managed read-only fields."""
+        payload = dict(workflow_json)
+        payload.pop("active", None)
+        payload.pop("id", None)
+        payload.pop("createdAt", None)
+        payload.pop("updatedAt", None)
+        payload.pop("versionId", None)
+        return payload
+
+    async def _update_workflow_via_put(self, workflow_id: str, workflow_json: dict) -> dict:
+        """Fallback update path for n8n instances that only support PUT."""
+        return await self._request("PUT", f"workflows/{workflow_id}", json=self._workflow_payload(workflow_json))
 
     async def update_workflow(self, workflow_id: str, workflow_json: dict) -> dict:
         """Update an existing workflow."""
-        return await self._request("PATCH", f"workflows/{workflow_id}", json=workflow_json)
+        result = await self._request("PATCH", f"workflows/{workflow_id}", json=self._workflow_payload(workflow_json))
+        if result.get("status_code") == 405:
+            fallback = await self._update_workflow_via_put(workflow_id, workflow_json)
+            if not fallback.get("error") and isinstance(fallback, dict):
+                fallback.setdefault("method", "PUT")
+            return fallback
+        return result
     
     async def enable_workflow(self, workflow_id: str) -> dict:
         """Enable a workflow."""
@@ -232,6 +252,11 @@ class N8NSetupAgent:
         if result.get("error"):
             fallback = await self._request("PATCH", f"workflows/{workflow_id}", json={"active": True})
             if fallback.get("error"):
+                if fallback.get("status_code") == 405:
+                    put_fallback = await self._request("PUT", f"workflows/{workflow_id}", json={"active": True})
+                    if put_fallback.get("error"):
+                        return put_fallback
+                    return {"success": True, "id": workflow_id, "method": "put"}
                 return fallback
             return {"success": True, "id": workflow_id, "method": "patch"}
         return {"success": True, "id": workflow_id, "method": "activate"}
@@ -250,12 +275,12 @@ class N8NSetupAgent:
                     return value
         return []
 
-    def build_telegram_workflow(self, webhook_path: str, backend_url: str) -> dict:
+    def build_telegram_workflow(self, webhook_path: str, backend_url: str, allowed_chat_id: str | None = None) -> dict:
         """Build the Telegram-to-backend workflow definition."""
         normalized_path = webhook_path.strip().strip("/") or "telegram"
+        allowed_chat = json.dumps(str(allowed_chat_id)) if allowed_chat_id else "''"
         return {
             "name": "Telegram to AI Software Factory",
-            "active": False,
             "settings": {"executionOrder": "v1"},
             "nodes": [
                 {
@@ -273,13 +298,13 @@ class N8NSetupAgent:
                 },
                 {
                     "id": "parse-node",
-                    "name": "Prepare Software Request",
+                    "name": "Prepare Freeform Request",
                     "type": "n8n-nodes-base.code",
                     "typeVersion": 2,
                     "position": [-200, 120],
                     "parameters": {
                         "language": "javaScript",
-                        "jsCode": "const body = $json.body ?? $json;\nconst message = body.message ?? body;\nconst text = String(message.text ?? '').trim();\nconst lines = text.split(/\\r?\\n/);\nconst request = { name: null, description: '', features: [], tech_stack: [] };\nlet nameIndex = -1;\nlet featuresIndex = -1;\nlet techIndex = -1;\nfor (let i = 0; i < lines.length; i += 1) {\n  const line = lines[i].trim();\n  if (line.toLowerCase().startsWith('name:')) { request.name = line.split(':', 2)[1]?.trim() || null; nameIndex = i; }\n  if (line.toLowerCase().startsWith('features:') && featuresIndex === -1) { featuresIndex = i; }\n  if (line.toLowerCase().startsWith('tech stack:') && techIndex === -1) { techIndex = i; }\n}\nif (nameIndex >= 0) {\n  const descriptionEnd = featuresIndex >= 0 ? featuresIndex : (techIndex >= 0 ? techIndex : lines.length);\n  request.description = lines.slice(nameIndex + 1, descriptionEnd).join('\\n').replace(/^description:\\s*/i, '').trim();\n}\nfunction collectList(startIndex, fieldName) {\n  if (startIndex < 0) return;\n  const firstLine = lines[startIndex].split(':').slice(1).join(':').trim();\n  if (firstLine && !firstLine.startsWith('-') && !firstLine.startsWith('*')) {\n    request[fieldName].push(...firstLine.split(',').map(item => item.trim()).filter(Boolean));\n  }\n  for (const rawLine of lines.slice(startIndex + 1)) {\n    const line = rawLine.trim();\n    if (!line) continue;\n    if (/^[A-Za-z ]+:/.test(line)) break;\n    if (line.startsWith('-') || line.startsWith('*')) {\n      const value = line.slice(1).trim();\n      if (value) request[fieldName].push(value);\n    }\n  }\n}\ncollectList(featuresIndex, 'features');\ncollectList(techIndex, 'tech_stack');\nif (!request.name || request.features.length === 0) { throw new Error('Could not parse software request from Telegram message'); }\nreturn [{ json: { ...request, _source: { raw_text: text, chat_id: message.chat?.id ?? null } } }];",
+                        "jsCode": f"const allowedChatId = {allowed_chat};\nconst body = $json.body ?? $json;\nconst message = body.message ?? body;\nconst text = String(message.text ?? '').trim();\nconst chatId = String(message.chat?.id ?? '');\nif (allowedChatId && chatId !== allowedChatId) {{\n  return [{{ json: {{ ignored: true, message: `Ignoring message from chat ${{chatId}}`, prompt_text: text, source: 'telegram', chat_id: chatId, chat_type: message.chat?.type ?? null }} }}];\n}}\nreturn [{{ json: {{ prompt_text: text, source: 'telegram', chat_id: chatId, chat_type: message.chat?.type ?? null }} }}];",
                     },
                 },
                 {
@@ -310,8 +335,8 @@ class N8NSetupAgent:
                 },
             ],
             "connections": {
-                "Telegram Webhook": {"main": [[{"node": "Prepare Software Request", "type": "main", "index": 0}]]},
-                "Prepare Software Request": {"main": [[{"node": "AI Software Factory API", "type": "main", "index": 0}]]},
+                "Telegram Webhook": {"main": [[{"node": "Prepare Freeform Request", "type": "main", "index": 0}]]},
+                "Prepare Freeform Request": {"main": [[{"node": "AI Software Factory API", "type": "main", "index": 0}]]},
                 "AI Software Factory API": {"main": [[{"node": "Respond to Telegram Webhook", "type": "main", "index": 0}]]},
             },
         }
@@ -320,11 +345,12 @@ class N8NSetupAgent:
         self,
         backend_url: str,
         credential_name: str,
+        allowed_chat_id: str | None = None,
     ) -> dict:
         """Build a production Telegram Trigger based workflow."""
+        allowed_chat = json.dumps(str(allowed_chat_id)) if allowed_chat_id else "''"
         return {
             "name": "Telegram to AI Software Factory",
-            "active": False,
             "settings": {"executionOrder": "v1"},
             "nodes": [
                 {
@@ -333,18 +359,18 @@ class N8NSetupAgent:
                     "type": "n8n-nodes-base.telegramTrigger",
                     "typeVersion": 1,
                     "position": [-520, 120],
-                    "parameters": {"updates": ["message"]},
+                    "parameters": {"updates": ["message", "channel_post"]},
                     "credentials": {"telegramApi": {"name": credential_name}},
                 },
                 {
-                    "id": "parse-node",
-                    "name": "Prepare Software Request",
+                    "id": "filter-node",
+                    "name": "Prepare Freeform Request",
                     "type": "n8n-nodes-base.code",
                     "typeVersion": 2,
                     "position": [-180, 120],
                     "parameters": {
                         "language": "javaScript",
-                        "jsCode": "const message = $json.message ?? $json;\nconst text = String(message.text ?? '').trim();\nconst lines = text.split(/\\r?\\n/);\nconst request = { name: null, description: '', features: [], tech_stack: [], _source: { raw_text: text, chat_id: message.chat?.id ?? null } };\nlet nameIndex = -1;\nlet featuresIndex = -1;\nlet techIndex = -1;\nfor (let i = 0; i < lines.length; i += 1) {\n  const line = lines[i].trim();\n  if (line.toLowerCase().startsWith('name:')) { request.name = line.split(':', 2)[1]?.trim() || null; nameIndex = i; }\n  if (line.toLowerCase().startsWith('features:') && featuresIndex === -1) { featuresIndex = i; }\n  if (line.toLowerCase().startsWith('tech stack:') && techIndex === -1) { techIndex = i; }\n}\nif (nameIndex >= 0) {\n  const descriptionEnd = featuresIndex >= 0 ? featuresIndex : (techIndex >= 0 ? techIndex : lines.length);\n  request.description = lines.slice(nameIndex + 1, descriptionEnd).join('\\n').replace(/^description:\\s*/i, '').trim();\n}\nfunction collectList(startIndex, fieldName) {\n  if (startIndex < 0) return;\n  const firstLine = lines[startIndex].split(':').slice(1).join(':').trim();\n  if (firstLine && !firstLine.startsWith('-') && !firstLine.startsWith('*')) {\n    request[fieldName].push(...firstLine.split(',').map(item => item.trim()).filter(Boolean));\n  }\n  for (const rawLine of lines.slice(startIndex + 1)) {\n    const line = rawLine.trim();\n    if (!line) continue;\n    if (/^[A-Za-z ]+:/.test(line)) break;\n    if (line.startsWith('-') || line.startsWith('*')) {\n      const value = line.slice(1).trim();\n      if (value) request[fieldName].push(value);\n    }\n  }\n}\ncollectList(featuresIndex, 'features');\ncollectList(techIndex, 'tech_stack');\nif (!request.name || request.features.length === 0) { throw new Error('Could not parse software request from Telegram message'); }\nreturn [{ json: request }];",
+                        "jsCode": f"const allowedChatId = {allowed_chat};\nconst message = $json.message ?? $json.channel_post ?? $json;\nconst text = String(message.text ?? '').trim();\nconst chatId = String(message.chat?.id ?? '');\nif (!text) return [];\nif (allowedChatId && chatId !== allowedChatId) return [];\nreturn [{{ json: {{ prompt_text: text, source: 'telegram', chat_id: chatId, chat_type: message.chat?.type ?? null }} }}];",
                     },
                 },
                 {
@@ -371,15 +397,15 @@ class N8NSetupAgent:
                     "parameters": {
                         "resource": "message",
                         "operation": "sendMessage",
-                        "chatId": "={{ $('Telegram Trigger').item.json.message.chat.id }}",
-                        "text": "={{ $json.data ? `Generated ${$json.data.name} (${($json.data.changed_files || []).length} files)` : ($json.message || 'Software generation request accepted') }}",
+                        "chatId": "={{ ($('Telegram Trigger').item.json.message ?? $('Telegram Trigger').item.json.channel_post).chat.id }}",
+                        "text": "={{ $json.summary_message || $json.data?.summary_message || $json.message || 'Software generation request accepted' }}",
                     },
                     "credentials": {"telegramApi": {"name": credential_name}},
                 },
             ],
             "connections": {
-                "Telegram Trigger": {"main": [[{"node": "Prepare Software Request", "type": "main", "index": 0}]]},
-                "Prepare Software Request": {"main": [[{"node": "AI Software Factory API", "type": "main", "index": 0}]]},
+                "Telegram Trigger": {"main": [[{"node": "Prepare Freeform Request", "type": "main", "index": 0}]]},
+                "Prepare Freeform Request": {"main": [[{"node": "AI Software Factory API", "type": "main", "index": 0}]]},
                 "AI Software Factory API": {"main": [[{"node": "Send Telegram Update", "type": "main", "index": 0}]]},
             },
         }
@@ -433,7 +459,7 @@ class N8NSetupAgent:
         """
         return await self.setup(
             webhook_path=webhook_path,
-            backend_url=f"{settings.backend_public_url}/generate",
+            backend_url=f"{settings.backend_public_url}/generate/text",
             force_update=False,
         )
     
@@ -470,7 +496,7 @@ class N8NSetupAgent:
                 "suggestion": health.get("suggestion"),
             }
         
-        effective_backend_url = backend_url or f"{settings.backend_public_url}/generate"
+        effective_backend_url = backend_url or f"{settings.backend_public_url}/generate/text"
         effective_bot_token = telegram_bot_token or settings.telegram_bot_token
         effective_credential_name = telegram_credential_name or settings.n8n_telegram_credential_name
         trigger_mode = use_telegram_trigger if use_telegram_trigger is not None else bool(effective_bot_token)
@@ -482,11 +508,13 @@ class N8NSetupAgent:
             workflow = self.build_telegram_trigger_workflow(
                 backend_url=effective_backend_url,
                 credential_name=effective_credential_name,
+                allowed_chat_id=settings.telegram_chat_id,
             )
         else:
             workflow = self.build_telegram_workflow(
                 webhook_path=webhook_path,
                 backend_url=effective_backend_url,
+                allowed_chat_id=settings.telegram_chat_id,
             )
 
         existing = await self.get_workflow(workflow["name"])

ai_software_factory/agents/orchestrator.py

@@ -2,7 +2,10 @@
 
 from __future__ import annotations
 
+import difflib
 import py_compile
+import re
+import subprocess
 from typing import Optional
 from datetime import datetime
 
@@ -33,6 +36,11 @@ class AgentOrchestrator:
         db=None,
         prompt_text: str | None = None,
         prompt_actor: str = "api",
+        existing_history=None,
+        prompt_source_context: dict | None = None,
+        prompt_routing: dict | None = None,
+        repo_name_override: str | None = None,
+        related_issue_hint: dict | None = None,
     ):
         """Initialize orchestrator."""
         self.project_id = project_id
@@ -49,6 +57,10 @@ class AgentOrchestrator:
         self.db = db
         self.prompt_text = prompt_text
         self.prompt_actor = prompt_actor
+        self.prompt_source_context = prompt_source_context or {}
+        self.prompt_routing = prompt_routing or {}
+        self.repo_name_override = repo_name_override
+        self.existing_history = existing_history
         self.changed_files: list[str] = []
         self.gitea_api = GiteaAPI(
             token=settings.GITEA_TOKEN,
@@ -58,12 +70,18 @@ class AgentOrchestrator:
         )
         self.project_root = settings.projects_root / project_id
         self.prompt_audit = None
-        self.repo_name = settings.gitea_repo or self.gitea_api.build_project_repo_name(project_id, project_name)
+        self.repo_name = settings.gitea_repo or self.gitea_api.build_project_repo_name(project_id, repo_name_override or project_name)
         self.repo_owner = settings.gitea_owner
-        self.repo_url = self._build_repo_url(self.repo_owner, self.repo_name)
+        self.repo_url = None
+        self.branch_name = self._build_pr_branch_name(project_id)
+        self.active_pull_request = None
+        self._gitea_username: str | None = None
+        hinted_issue_number = (related_issue_hint or {}).get('number') if related_issue_hint else None
+        self.related_issue_number = hinted_issue_number if hinted_issue_number is not None else self._extract_issue_number(prompt_text)
+        self.related_issue: dict | None = DatabaseManager._normalize_issue(related_issue_hint)
 
         # Initialize agents
-        self.git_manager = GitManager(project_id)
+        self.git_manager = GitManager(project_id, project_dir=str(self.project_root))
         self.ui_manager = UIManager(project_id)
 
         # Initialize database manager if db session provided
@@ -71,22 +89,37 @@ class AgentOrchestrator:
         self.history = None
         if db:
             self.db_manager = DatabaseManager(db)
-            # Log project start to database
-            self.history = self.db_manager.log_project_start(
-                project_id=project_id,
-                project_name=project_name,
-                description=description
-            )
-            # Re-fetch with new history_id
-            self.db_manager = DatabaseManager(db)
+            if existing_history is not None:
+                self.history = existing_history
+                self.project_id = existing_history.project_id
+                self.project_name = existing_history.project_name or project_name
+                self.description = existing_history.description or description
+            else:
+                self.history = self.db_manager.log_project_start(
+                    project_id=project_id,
+                    project_name=project_name,
+                    description=description
+                )
+                self.db_manager = DatabaseManager(db)
+            self.active_pull_request = self.db_manager.get_open_pull_request(project_id=self.project_id)
+            if existing_history is not None and self.history is not None:
+                latest_ui = self.db_manager._get_latest_ui_snapshot_data(self.history.id)
+                repository = latest_ui.get('repository') if isinstance(latest_ui, dict) else None
+                if isinstance(repository, dict) and repository:
+                    self.repo_owner = repository.get('owner') or self.repo_owner
+                    self.repo_name = repository.get('name') or self.repo_name
+                    self.repo_url = repository.get('url') or self.repo_url
             if self.prompt_text:
                 self.prompt_audit = self.db_manager.log_prompt_submission(
                     history_id=self.history.id,
-                    project_id=project_id,
+                    project_id=self.project_id,
                     prompt_text=self.prompt_text,
                     features=self.features,
                     tech_stack=self.tech_stack,
                     actor_name=self.prompt_actor,
+                    related_issue={'number': self.related_issue_number} if self.related_issue_number is not None else None,
+                    source_context=self.prompt_source_context,
+                    routing=self.prompt_routing,
                 )
 
         self.ui_manager.ui_data["project_root"] = str(self.project_root)
@@ -95,19 +128,108 @@ class AgentOrchestrator:
         self.ui_manager.ui_data["repository"] = {
             "owner": self.repo_owner,
             "name": self.repo_name,
-            "url": self.repo_url,
             "mode": "project" if settings.use_project_repositories else "shared",
+            "status": "pending" if settings.use_project_repositories else "shared",
+            "provider": "gitea",
         }
+        if self.related_issue:
+            self.ui_manager.ui_data["related_issue"] = self.related_issue
+        if self.active_pull_request:
+            self.ui_manager.ui_data["pull_request"] = self.active_pull_request
+
+    def _build_pr_branch_name(self, project_id: str) -> str:
+        """Build a stable branch name used until the PR is merged."""
+        return f"ai/{project_id}"
+
+    def _extract_issue_number(self, prompt_text: str | None) -> int | None:
+        """Extract an issue reference from prompt text."""
+        if not prompt_text:
+            return None
+        match = re.search(r'(?:#|issue\s+)(\d+)', prompt_text, flags=re.IGNORECASE)
+        return int(match.group(1)) if match else None
 
     def _build_repo_url(self, owner: str | None, repo: str | None) -> str | None:
         if not owner or not repo or not settings.gitea_url:
             return None
         return f"{settings.gitea_url.rstrip('/')}/{owner}/{repo}"
 
+    def _log_generation_plan_trace(self) -> None:
+        """Persist the current generation plan as an inspectable trace."""
+        if not self.db_manager or not self.history or not self.prompt_audit:
+            return
+        planned_files = list(self._template_files().keys())
+        self.db_manager.log_llm_trace(
+            project_id=self.project_id,
+            history_id=self.history.id,
+            prompt_id=self.prompt_audit.id,
+            stage='generation_plan',
+            provider='factory-planner',
+            model='template-generator',
+            system_prompt='Plan the generated project structure from the structured request and repository state.',
+            user_prompt=self.prompt_text or self.description,
+            assistant_response=(
+                f"Planned files: {', '.join(planned_files)}. "
+                f"Target branch: {self.branch_name}. "
+                f"Repository mode: {self.ui_manager.ui_data.get('repository', {}).get('mode', 'unknown')}."
+                + (
+                    f" Linked issue: #{self.related_issue.get('number')} {self.related_issue.get('title')}."
+                    if self.related_issue else ''
+                )
+            ),
+            raw_response={
+                'planned_files': planned_files,
+                'features': list(self.features),
+                'tech_stack': list(self.tech_stack),
+                'branch': self.branch_name,
+                'repository': self.ui_manager.ui_data.get('repository', {}),
+                'related_issue': self.related_issue,
+            },
+            fallback_used=False,
+        )
+
+    async def _sync_issue_context(self) -> None:
+        """Sync repository issues and resolve a linked issue from the prompt when present."""
+        if not self.db_manager or not self.history:
+            return
+        repository = self.ui_manager.ui_data.get('repository') or {}
+        owner = repository.get('owner') or self.repo_owner
+        repo_name = repository.get('name') or self.repo_name
+        if not owner or not repo_name or not settings.gitea_url or not settings.gitea_token:
+            return
+        issues_result = self.db_manager.sync_repository_issues(project_id=self.project_id, gitea_api=self.gitea_api, state='open')
+        self.ui_manager.ui_data['issues'] = issues_result.get('issues', []) if issues_result.get('status') == 'success' else []
+        if self.related_issue_number is None:
+            return
+        issue_payload = await self.gitea_api.get_issue(issue_number=self.related_issue_number, owner=owner, repo=repo_name)
+        if isinstance(issue_payload, dict) and issue_payload.get('error'):
+            return
+        if issue_payload.get('pull_request'):
+            return
+        self.related_issue = DatabaseManager._normalize_issue(issue_payload)
+        self.ui_manager.ui_data['related_issue'] = self.related_issue
+        if self.prompt_audit:
+            self.db_manager.attach_issue_to_prompt(self.prompt_audit.id, self.related_issue)
+
     async def _ensure_remote_repository(self) -> None:
         if not settings.use_project_repositories:
+            self.ui_manager.ui_data["repository"]["status"] = "shared"
+            if settings.gitea_repo:
+                predicted_url = self._build_repo_url(self.repo_owner, self.repo_name)
+                if predicted_url:
+                    self.repo_url = predicted_url
+                    self.ui_manager.ui_data["repository"]["url"] = predicted_url
+            self.ui_manager.ui_data["repository"]["api_response"] = {
+                "status": "shared",
+                "detail": "Using the configured shared repository instead of provisioning a per-project repo.",
+            }
             return
         if not self.repo_owner or not settings.gitea_token or not settings.gitea_url:
+            self.ui_manager.ui_data["repository"]["status"] = "skipped"
+            self.ui_manager.ui_data["repository"]["reason"] = "Missing Gitea owner, URL, or token configuration"
+            self.ui_manager.ui_data["repository"]["api_response"] = {
+                "status": "skipped",
+                "detail": "Missing Gitea owner, URL, or token configuration",
+            }
             return
 
         repo_name = self.repo_name
@@ -115,6 +237,7 @@ class AgentOrchestrator:
             repo_name=repo_name,
             owner=self.repo_owner,
             description=f"AI-generated project for {self.project_name}",
+            auto_init=False,
         )
         if result.get("status") == "exists" and repo_name == self.gitea_api.build_project_repo_name(self.project_id, self.project_name):
             repo_name = f"{repo_name}-{self.project_id.split('-')[-1]}"
@@ -122,6 +245,7 @@ class AgentOrchestrator:
                 repo_name=repo_name,
                 owner=self.repo_owner,
                 description=f"AI-generated project for {self.project_name}",
+                auto_init=False,
             )
             self.repo_name = repo_name
             self.ui_manager.ui_data["repository"]["name"] = repo_name
@@ -135,10 +259,176 @@ class AgentOrchestrator:
                     else f"Prepared repository {self.repo_owner}/{self.repo_name}"
                 ),
             )
-        self.ui_manager.ui_data["repository"]["status"] = result.get("status", "error" if result.get("error") else "ready")
-        if result.get("html_url"):
+        repo_status = result.get("status", "error" if result.get("error") else "ready")
+        self.ui_manager.ui_data["repository"]["status"] = repo_status
+        self.ui_manager.ui_data["repository"]["api_response"] = {
+            key: value
+            for key, value in result.items()
+            if key not in {"private"}
+        }
+        if result.get("status_code") is not None:
+            self.ui_manager.ui_data["repository"]["api_status_code"] = result.get("status_code")
+        if result.get("error"):
+            self.ui_manager.ui_data["repository"]["reason"] = result.get("error")
+            self.ui_manager.ui_data["repository"].pop("url", None)
+        elif result.get("html_url"):
             self.repo_url = result["html_url"]
             self.ui_manager.ui_data["repository"]["url"] = self.repo_url
+            clone_url = result.get("clone_url") or self.gitea_api.build_repo_git_url(self.repo_owner, self.repo_name)
+            if clone_url:
+                self.ui_manager.ui_data["repository"]["clone_url"] = clone_url
+            self.ui_manager.ui_data["repository"].pop("reason", None)
+        elif repo_status == "exists":
+            predicted_url = self._build_repo_url(self.repo_owner, self.repo_name)
+            if predicted_url:
+                self.repo_url = predicted_url
+                self.ui_manager.ui_data["repository"]["url"] = predicted_url
+            clone_url = result.get("clone_url") or self.gitea_api.build_repo_git_url(self.repo_owner, self.repo_name)
+            if clone_url:
+                self.ui_manager.ui_data["repository"]["clone_url"] = clone_url
+        else:
+            self.ui_manager.ui_data["repository"].pop("url", None)
+
+    async def _resolve_gitea_username(self) -> str:
+        """Resolve and cache the Gitea login used for authenticated git operations."""
+        if self._gitea_username:
+            return self._gitea_username
+        user_info = await self.gitea_api.get_current_user()
+        if user_info.get('error') or not user_info.get('login'):
+            raise RuntimeError(f"Unable to resolve Gitea user for push: {user_info.get('error', 'missing login')}")
+        self._gitea_username = user_info['login']
+        return self._gitea_username
+
+    async def _push_branch(self, branch: str) -> dict | None:
+        """Push a branch to the configured project repository when available."""
+        repository = self.ui_manager.ui_data.get('repository') or {}
+        if repository.get('mode') != 'project':
+            return None
+        if repository.get('status') not in {'created', 'exists', 'ready'}:
+            return None
+        if not settings.gitea_token or not self.repo_owner or not self.repo_name:
+            return None
+
+        clone_url = repository.get('clone_url') or self.gitea_api.build_repo_git_url(self.repo_owner, self.repo_name)
+        if not clone_url:
+            return None
+        username = await self._resolve_gitea_username()
+        self.git_manager.push_with_credentials(
+            remote_url=clone_url,
+            username=username,
+            password=settings.gitea_token,
+            remote='origin',
+            branch=branch,
+        )
+        return {'status': 'pushed', 'remote': clone_url, 'branch': branch}
+
+    async def _prepare_git_workspace(self) -> None:
+        """Initialize the local repo and ensure the PR branch exists before writing files."""
+        if not self.git_manager.is_git_available():
+            self.ui_manager.ui_data.setdefault('git', {})['error'] = 'git executable is not available in PATH'
+            self._append_log('Local git workspace skipped: git executable is not available in PATH')
+            return
+        if not self.git_manager.has_repo():
+            self.git_manager.init_repo()
+
+        if not self.git_manager.current_head_or_none():
+            self.git_manager.create_empty_commit('Initialize project repository')
+            try:
+                await self._push_branch('main')
+            except (RuntimeError, subprocess.CalledProcessError, FileNotFoundError) as exc:
+                self.ui_manager.ui_data.setdefault('git', {})['remote_error'] = str(exc)
+                self._append_log(f'Initial main push skipped: {exc}')
+
+        if self.git_manager.branch_exists(self.branch_name):
+            self.git_manager.checkout_branch(self.branch_name)
+        else:
+            self.git_manager.checkout_branch(self.branch_name, create=True, start_point='main')
+        self.ui_manager.ui_data.setdefault('git', {})['active_branch'] = self.branch_name
+
+    async def _ensure_pull_request(self) -> dict | None:
+        """Create the project pull request on first delivery and reuse it later."""
+        if self.active_pull_request:
+            self.ui_manager.ui_data['pull_request'] = self.active_pull_request
+            return self.active_pull_request
+        repository = self.ui_manager.ui_data.get('repository') or {}
+        if repository.get('mode') != 'project' or repository.get('status') not in {'created', 'exists', 'ready'}:
+            return None
+
+        title = f"AI delivery for {self.project_name}"
+        body = (
+            f"Automated software factory changes for {self.project_name}.\n\n"
+            f"Prompt: {self.prompt_text or self.description}\n\n"
+            f"Branch: {self.branch_name}"
+        )
+        result = await self.gitea_api.create_pull_request(
+            title=title,
+            body=body,
+            owner=self.repo_owner,
+            repo=self.repo_name,
+            base='main',
+            head=self.branch_name,
+        )
+        if result.get('error'):
+            raise RuntimeError(f"Unable to create pull request: {result.get('error')}")
+
+        pr_number = result.get('number') or result.get('id') or 0
+        pr_data = {
+            'pr_number': pr_number,
+            'title': result.get('title', title),
+            'body': result.get('body', body),
+            'state': result.get('state', 'open'),
+            'base': result.get('base', {}).get('ref', 'main') if isinstance(result.get('base'), dict) else 'main',
+            'user': result.get('user', {}).get('login', 'system') if isinstance(result.get('user'), dict) else 'system',
+            'pr_url': result.get('html_url') or self.gitea_api.build_pull_request_url(pr_number, self.repo_owner, self.repo_name),
+            'merged': bool(result.get('merged')),
+            'pr_state': result.get('state', 'open'),
+        }
+        if self.db_manager and self.history:
+            self.db_manager.save_pr_data(self.history.id, pr_data)
+        self.active_pull_request = self.db_manager.get_open_pull_request(project_id=self.project_id) if self.db_manager else pr_data
+        self.ui_manager.ui_data['pull_request'] = self.active_pull_request or pr_data
+        return self.active_pull_request or pr_data
+
+    async def _push_remote_commit(self, commit_hash: str, commit_message: str, changed_files: list[str], base_commit: str | None) -> dict | None:
+        """Push the local commit to the provisioned Gitea repository and build browser links."""
+        repository = self.ui_manager.ui_data.get("repository") or {}
+        if repository.get("mode") != "project":
+            return None
+        if repository.get("status") not in {"created", "exists", "ready"}:
+            return None
+        push_result = await self._push_branch(self.branch_name)
+        if push_result is None:
+            return None
+        pull_request = await self._ensure_pull_request()
+        commit_url = self.gitea_api.build_commit_url(commit_hash, owner=self.repo_owner, repo=self.repo_name)
+        compare_url = self.gitea_api.build_compare_url(base_commit, commit_hash, owner=self.repo_owner, repo=self.repo_name) if base_commit else None
+        remote_record = {
+            "status": "pushed",
+            "remote": push_result.get('remote'),
+            "branch": self.branch_name,
+            "commit_url": commit_url,
+            "compare_url": compare_url,
+            "changed_files": changed_files,
+            "pull_request": pull_request,
+        }
+        self.ui_manager.ui_data.setdefault("git", {})["remote_push"] = remote_record
+        repository["last_commit_url"] = commit_url
+        if compare_url:
+            repository["last_compare_url"] = compare_url
+        self._append_log(f"Pushed generated commit to {self.repo_owner}/{self.repo_name}.")
+        return remote_record
+
+    def _build_diff_text(self, relative_path: str, previous_content: str, new_content: str) -> str:
+        """Build a unified diff for display in the dashboard."""
+        previous_lines = previous_content.splitlines(keepends=True)
+        new_lines = new_content.splitlines(keepends=True)
+        diff = difflib.unified_diff(
+            previous_lines,
+            new_lines,
+            fromfile=f"a/{relative_path}",
+            tofile=f"b/{relative_path}",
+        )
+        return "".join(diff)
 
     def _append_log(self, message: str) -> None:
         timestamped = f"[{datetime.utcnow().isoformat()}] {message}"
@@ -163,6 +453,8 @@ class AgentOrchestrator:
         target = self.project_root / relative_path
         target.parent.mkdir(parents=True, exist_ok=True)
         change_type = "UPDATE" if target.exists() else "CREATE"
+        previous_content = target.read_text(encoding="utf-8") if target.exists() else ""
+        diff_text = self._build_diff_text(relative_path, previous_content, content)
         target.write_text(content, encoding="utf-8")
         self.changed_files.append(relative_path)
         if self.db_manager and self.history:
@@ -176,6 +468,7 @@ class AgentOrchestrator:
                 history_id=self.history.id,
                 prompt_id=self.prompt_audit.id if self.prompt_audit else None,
                 diff_summary=f"Wrote {len(content.splitlines())} lines to {relative_path}",
+                diff_text=diff_text,
             )
 
     def _template_files(self) -> dict[str, str]:
@@ -215,6 +508,9 @@ class AgentOrchestrator:
             self._append_log("Initializing project.")
 
             await self._ensure_remote_repository()
+            await self._sync_issue_context()
+            await self._prepare_git_workspace()
+            self._log_generation_plan_trace()
 
             # Step 2: Create project structure (skip git operations)
             self._update_progress(20, "project-structure", "Creating project files...")
@@ -228,6 +524,10 @@ class AgentOrchestrator:
             self._update_progress(80, "validation", "Validating generated code...")
             await self._run_tests()
 
+            # Step 5: Commit generated artifacts locally for traceability
+            self._update_progress(90, "git", "Recording generated changes in git...")
+            await self._commit_to_git()
+
             # Step 7: Complete
             self.status = "completed"
             self._update_progress(100, "completed", "Software generation complete!")
@@ -253,6 +553,8 @@ class AgentOrchestrator:
                 "project_root": str(self.project_root),
                 "changed_files": list(dict.fromkeys(self.changed_files)),
                 "repository": self.ui_manager.ui_data.get("repository"),
+                "related_issue": self.related_issue,
+                "pull_request": self.ui_manager.ui_data.get("pull_request"),
             }
 
         except Exception as e:
@@ -279,6 +581,8 @@ class AgentOrchestrator:
                 "project_root": str(self.project_root),
                 "changed_files": list(dict.fromkeys(self.changed_files)),
                 "repository": self.ui_manager.ui_data.get("repository"),
+                "related_issue": self.related_issue,
+                "pull_request": self.ui_manager.ui_data.get("pull_request"),
             }
 
     async def _create_project_structure(self) -> None:
@@ -305,7 +609,78 @@ class AgentOrchestrator:
 
     async def _commit_to_git(self) -> None:
         """Commit changes to git."""
-        pass  # Skip git operations in test environment
+        unique_files = list(dict.fromkeys(self.changed_files))
+        if not unique_files:
+            return
+        if not self.git_manager.is_git_available():
+            self.ui_manager.ui_data.setdefault('git', {})['error'] = 'git executable is not available in PATH'
+            self._append_log('Git commit skipped: git executable is not available in PATH')
+            return
+
+        try:
+            if not self.git_manager.has_repo():
+                self.git_manager.init_repo()
+            base_commit = self.git_manager.current_head_or_none()
+            self.git_manager.add_files(unique_files)
+            if not self.git_manager.get_status():
+                return
+
+            commit_message = f"AI generation for prompt: {self.project_name}"
+            commit_hash = self.git_manager.commit(commit_message)
+            commit_record = {
+                "hash": commit_hash,
+                "message": commit_message,
+                "files": unique_files,
+                "timestamp": datetime.utcnow().isoformat(),
+                "scope": "local",
+                "branch": self.branch_name,
+            }
+            remote_record = None
+            try:
+                remote_record = await self._push_remote_commit(commit_hash, commit_message, unique_files, base_commit)
+            except (RuntimeError, subprocess.CalledProcessError, FileNotFoundError) as remote_exc:
+                self.ui_manager.ui_data.setdefault("git", {})["remote_error"] = str(remote_exc)
+                self._append_log(f"Remote git push skipped: {remote_exc}")
+
+            if remote_record:
+                commit_record["scope"] = "remote"
+                commit_record["commit_url"] = remote_record.get("commit_url")
+                commit_record["compare_url"] = remote_record.get("compare_url")
+                if remote_record.get('pull_request'):
+                    commit_record['pull_request'] = remote_record['pull_request']
+                    self.ui_manager.ui_data['pull_request'] = remote_record['pull_request']
+            self.ui_manager.ui_data.setdefault("git", {})["latest_commit"] = commit_record
+            self.ui_manager.ui_data.setdefault("git", {})["commits"] = [commit_record]
+            self._append_log(f"Recorded git commit {commit_hash[:12]} for generated files.")
+            if self.db_manager:
+                self.db_manager.log_commit(
+                    project_id=self.project_id,
+                    commit_message=commit_message,
+                    actor="orchestrator",
+                    actor_type="agent",
+                    history_id=self.history.id if self.history else None,
+                    prompt_id=self.prompt_audit.id if self.prompt_audit else None,
+                    commit_hash=commit_hash,
+                    changed_files=unique_files,
+                    branch=self.branch_name,
+                    commit_url=remote_record.get("commit_url") if remote_record else None,
+                    compare_url=remote_record.get("compare_url") if remote_record else None,
+                    remote_status=remote_record.get("status") if remote_record else "local-only",
+                    related_issue=self.related_issue,
+                )
+                if self.related_issue:
+                    self.db_manager.log_issue_work(
+                        project_id=self.project_id,
+                        history_id=self.history.id if self.history else None,
+                        prompt_id=self.prompt_audit.id if self.prompt_audit else None,
+                        issue=self.related_issue,
+                        actor='orchestrator',
+                        commit_hash=commit_hash,
+                        commit_url=remote_record.get('commit_url') if remote_record else None,
+                    )
+        except (RuntimeError, subprocess.CalledProcessError, FileNotFoundError) as exc:
+            self.ui_manager.ui_data.setdefault("git", {})["error"] = str(exc)
+            self._append_log(f"Git commit skipped: {exc}")
 
     async def _create_pr(self) -> None:
         """Create pull request."""

ai_software_factory/agents/prompt_workflow.py

@@ -0,0 +1,127 @@
+"""Helpers for prompt-level repository workflows such as undoing a prompt."""
+
+from __future__ import annotations
+
+import subprocess
+
+try:
+    from ..config import settings
+    from .database_manager import DatabaseManager
+    from .git_manager import GitManager
+    from .gitea import GiteaAPI
+except ImportError:
+    from config import settings
+    from agents.database_manager import DatabaseManager
+    from agents.git_manager import GitManager
+    from agents.gitea import GiteaAPI
+
+
+class PromptWorkflowManager:
+    """Coordinate prompt-level repository actions against git and Gitea."""
+
+    def __init__(self, db):
+        self.db_manager = DatabaseManager(db)
+        self.gitea_api = GiteaAPI(
+            token=settings.GITEA_TOKEN,
+            base_url=settings.GITEA_URL,
+            owner=settings.GITEA_OWNER,
+            repo=settings.GITEA_REPO or '',
+        )
+
+    async def undo_prompt(self, project_id: str, prompt_id: int) -> dict:
+        """Revert the commit associated with a prompt and push the revert to the PR branch."""
+        history = self.db_manager.get_project_by_id(project_id)
+        if history is None:
+            return {'status': 'error', 'message': 'Project not found'}
+
+        correlations = self.db_manager.get_prompt_change_correlations(project_id=project_id, limit=500)
+        correlation = next((item for item in correlations if item.get('prompt_id') == prompt_id), None)
+        if correlation is None:
+            return {'status': 'error', 'message': 'Prompt not found for project'}
+        if correlation.get('revert'):
+            return {'status': 'ignored', 'message': 'Prompt has already been reverted', 'revert': correlation['revert']}
+
+        original_commit = next(
+            (commit for commit in correlation.get('commits', []) if commit.get('remote_status') != 'reverted' and commit.get('commit_hash')),
+            None,
+        )
+        if original_commit is None:
+            return {'status': 'error', 'message': 'No reversible commit was recorded for this prompt'}
+
+        branch = original_commit.get('branch') or f'ai/{project_id}'
+        project_root = settings.projects_root / project_id
+        git_manager = GitManager(project_id, project_dir=str(project_root))
+        if not git_manager.has_repo():
+            return {'status': 'error', 'message': 'Local project repository is not available for undo'}
+
+        try:
+            git_manager.checkout_branch(branch)
+            previous_head = git_manager.current_head_or_none()
+            revert_commit_hash = git_manager.revert_commit(original_commit['commit_hash'])
+        except (subprocess.CalledProcessError, FileNotFoundError) as exc:
+            return {'status': 'error', 'message': f'Unable to revert prompt commit: {exc}'}
+
+        repository = self.db_manager.get_project_audit_data(project_id).get('repository') or {}
+        commit_url = None
+        compare_url = None
+        if (
+            repository.get('mode') == 'project'
+            and repository.get('status') in {'created', 'exists', 'ready'}
+            and settings.gitea_token
+            and repository.get('owner')
+            and repository.get('name')
+        ):
+            try:
+                user_info = await self.gitea_api.get_current_user()
+                username = user_info.get('login') if isinstance(user_info, dict) else None
+                if username and not user_info.get('error'):
+                    remote_url = repository.get('clone_url') or self.gitea_api.build_repo_git_url(repository.get('owner'), repository.get('name'))
+                    if remote_url:
+                        git_manager.push_with_credentials(
+                            remote_url=remote_url,
+                            username=username,
+                            password=settings.gitea_token,
+                            branch=branch,
+                        )
+                        commit_url = self.gitea_api.build_commit_url(revert_commit_hash, repository.get('owner'), repository.get('name'))
+                        if previous_head:
+                            compare_url = self.gitea_api.build_compare_url(previous_head, revert_commit_hash, repository.get('owner'), repository.get('name'))
+            except (RuntimeError, subprocess.CalledProcessError, FileNotFoundError):
+                pass
+
+        self.db_manager.log_commit(
+            project_id=project_id,
+            commit_message=f'Revert prompt {prompt_id}',
+            actor='dashboard',
+            actor_type='user',
+            history_id=history.id,
+            prompt_id=prompt_id,
+            commit_hash=revert_commit_hash,
+            changed_files=original_commit.get('changed_files', []),
+            branch=branch,
+            commit_url=commit_url,
+            compare_url=compare_url,
+            remote_status='reverted',
+        )
+        self.db_manager.log_prompt_revert(
+            project_id=project_id,
+            prompt_id=prompt_id,
+            reverted_commit_hash=original_commit['commit_hash'],
+            revert_commit_hash=revert_commit_hash,
+            actor='dashboard',
+            commit_url=commit_url,
+        )
+        self.db_manager.log_system_event(
+            component='git',
+            level='INFO',
+            message=f'Reverted prompt {prompt_id} for project {project_id}',
+        )
+        return {
+            'status': 'success',
+            'project_id': project_id,
+            'prompt_id': prompt_id,
+            'reverted_commit_hash': original_commit['commit_hash'],
+            'revert_commit_hash': revert_commit_hash,
+            'commit_url': commit_url,
+            'compare_url': compare_url,
+        }

ai_software_factory/agents/request_interpreter.py

@@ -0,0 +1,395 @@
+"""Interpret free-form software requests into structured generation input."""
+
+from __future__ import annotations
+
+import json
+import re
+
+try:
+    from ..config import settings
+    from .gitea import GiteaAPI
+    from .llm_service import LLMServiceClient
+except ImportError:
+    from config import settings
+    from agents.gitea import GiteaAPI
+    from agents.llm_service import LLMServiceClient
+
+
+class RequestInterpreter:
+    """Use Ollama to turn free-form text into a structured software request."""
+
+    def __init__(self, ollama_url: str | None = None, model: str | None = None):
+        self.ollama_url = (ollama_url or settings.ollama_url).rstrip('/')
+        self.model = model or settings.OLLAMA_MODEL
+        self.llm_client = LLMServiceClient(ollama_url=self.ollama_url, model=self.model)
+        self.gitea_api = None
+        if settings.gitea_url and settings.gitea_token:
+            self.gitea_api = GiteaAPI(
+                token=settings.GITEA_TOKEN,
+                base_url=settings.GITEA_URL,
+                owner=settings.GITEA_OWNER,
+                repo=settings.GITEA_REPO or '',
+            )
+
+    async def interpret(self, prompt_text: str, context: dict | None = None) -> dict:
+        """Interpret free-form text into the request shape expected by the orchestrator."""
+        interpreted, _trace = await self.interpret_with_trace(prompt_text, context=context)
+        return interpreted
+
+    async def interpret_with_trace(self, prompt_text: str, context: dict | None = None) -> tuple[dict, dict]:
+        """Interpret free-form text into the request shape expected by the orchestrator."""
+        normalized = prompt_text.strip()
+        if not normalized:
+            raise ValueError('Prompt text cannot be empty')
+
+        compact_context = self._build_compact_context(context or {})
+
+        system_prompt = (
+            'You route Telegram software prompts. '
+            'Decide whether the prompt starts a new project or continues an existing tracked project. '
+            'When continuing, identify the best matching project_id from the provided context and the issue number if one is mentioned or implied by recent chat history. '
+            'Return only JSON with keys request and routing. '
+            'request must contain name, description, features, tech_stack. '
+            'routing must contain intent, project_id, project_name, issue_number, confidence, and reasoning_summary. '
+            'Use the provided project catalog and recent chat history. '
+            'If the user says things like also, continue, work on this, that issue, or follow-up wording, prefer continuation of the most relevant recent project. '
+            'If the user explicitly asks for a new project, set intent to new_project.'
+        )
+        user_prompt = normalized
+        if compact_context:
+            user_prompt = (
+                f"Conversation context:\n{json.dumps(compact_context, indent=2)}\n\n"
+                f"User prompt:\n{normalized}"
+            )
+
+        content, trace = await self.llm_client.chat_with_trace(
+            stage='request_interpretation',
+            system_prompt=system_prompt,
+            user_prompt=user_prompt,
+            tool_context_input={
+                'projects': compact_context.get('projects', []),
+                'open_issues': [
+                    issue
+                    for project in compact_context.get('projects', [])
+                    for issue in project.get('open_issues', [])
+                ],
+                'recent_chat_history': compact_context.get('recent_chat_history', []),
+            },
+            expect_json=True,
+        )
+        if content:
+            try:
+                parsed = json.loads(content)
+                interpreted = self._normalize_interpreted_request(parsed, normalized)
+                routing = self._normalize_routing(parsed.get('routing'), interpreted, compact_context)
+                naming_trace = None
+                if routing.get('intent') == 'new_project':
+                    interpreted, routing, naming_trace = await self._refine_new_project_identity(
+                        prompt_text=normalized,
+                        interpreted=interpreted,
+                        routing=routing,
+                        context=compact_context,
+                    )
+                trace['routing'] = routing
+                trace['context_excerpt'] = compact_context
+                if naming_trace is not None:
+                    trace['project_naming'] = naming_trace
+                return interpreted, trace
+            except Exception:
+                pass
+
+        interpreted, routing = self._heuristic_fallback(normalized, compact_context)
+        if routing.get('intent') == 'new_project':
+            constraints = await self._collect_project_identity_constraints(compact_context)
+            routing['repo_name'] = self._ensure_unique_repo_name(routing.get('repo_name') or interpreted.get('name') or 'project', constraints['repo_names'])
+        return interpreted, {
+            'stage': 'request_interpretation',
+            'provider': 'heuristic',
+            'model': self.model,
+            'system_prompt': system_prompt,
+            'user_prompt': user_prompt,
+            'assistant_response': json.dumps({'request': interpreted, 'routing': routing}),
+            'raw_response': {'fallback': 'heuristic', 'llm_trace': trace.get('raw_response') if isinstance(trace, dict) else None},
+            'routing': routing,
+            'context_excerpt': compact_context,
+            'guardrails': trace.get('guardrails') if isinstance(trace, dict) else [],
+            'tool_context': trace.get('tool_context') if isinstance(trace, dict) else [],
+            'fallback_used': True,
+        }
+
+    async def _refine_new_project_identity(
+        self,
+        *,
+        prompt_text: str,
+        interpreted: dict,
+        routing: dict,
+        context: dict,
+    ) -> tuple[dict, dict, dict | None]:
+        """Refine project and repository naming for genuinely new work."""
+        constraints = await self._collect_project_identity_constraints(context)
+        user_prompt = (
+            f"Original user prompt:\n{prompt_text}\n\n"
+            f"Draft structured request:\n{json.dumps(interpreted, indent=2)}\n\n"
+            f"Tracked project names to avoid reusing unless the user clearly wants them:\n{json.dumps(sorted(constraints['project_names']))}\n\n"
+            f"Repository slugs already reserved in tracked projects or Gitea:\n{json.dumps(sorted(constraints['repo_names']))}\n\n"
+            "Suggest the best project display name and repository slug for this new project."
+        )
+        content, trace = await self.llm_client.chat_with_trace(
+            stage='project_naming',
+            system_prompt=settings.llm_project_naming_system_prompt,
+            user_prompt=user_prompt,
+            tool_context_input={
+                'projects': context.get('projects', []),
+            },
+            expect_json=True,
+        )
+        if content:
+            try:
+                parsed = json.loads(content)
+                project_name, repo_name = self._normalize_project_identity(
+                    parsed,
+                    fallback_name=interpreted.get('name') or self._derive_name(prompt_text),
+                )
+                repo_name = self._ensure_unique_repo_name(repo_name, constraints['repo_names'])
+                interpreted['name'] = project_name
+                routing['project_name'] = project_name
+                routing['repo_name'] = repo_name
+                return interpreted, routing, trace
+            except Exception:
+                pass
+
+        fallback_name = interpreted.get('name') or self._derive_name(prompt_text)
+        routing['project_name'] = fallback_name
+        routing['repo_name'] = self._ensure_unique_repo_name(self._derive_repo_name(fallback_name), constraints['repo_names'])
+        return interpreted, routing, trace
+
+    async def _collect_project_identity_constraints(self, context: dict) -> dict[str, set[str]]:
+        """Collect reserved project names and repository slugs from tracked state and Gitea."""
+        project_names: set[str] = set()
+        repo_names: set[str] = set()
+        for project in context.get('projects', []):
+            if project.get('name'):
+                project_names.add(str(project.get('name')).strip())
+            repository = project.get('repository') or {}
+            if repository.get('name'):
+                repo_names.add(str(repository.get('name')).strip())
+        repo_names.update(await self._load_remote_repo_names())
+        return {
+            'project_names': project_names,
+            'repo_names': repo_names,
+        }
+
+    async def _load_remote_repo_names(self) -> set[str]:
+        """Load current Gitea repository names when live credentials are available."""
+        if settings.gitea_repo:
+            return {settings.gitea_repo}
+        if self.gitea_api is None or not settings.gitea_owner:
+            return set()
+        repos = await self.gitea_api.list_repositories(owner=settings.gitea_owner)
+        if not isinstance(repos, list):
+            return set()
+        return {str(repo.get('name')).strip() for repo in repos if repo.get('name')}
+
+    def _normalize_interpreted_request(self, interpreted: dict, original_prompt: str) -> dict:
+        """Normalize LLM output into the required request shape."""
+        request_payload = interpreted.get('request') if isinstance(interpreted.get('request'), dict) else interpreted
+        name = str(interpreted.get('name') or '').strip() or self._derive_name(original_prompt)
+        if isinstance(request_payload, dict):
+            name = str(request_payload.get('name') or '').strip() or self._derive_name(original_prompt)
+        description = str((request_payload or {}).get('description') or '').strip() or original_prompt[:255]
+        features = self._normalize_list((request_payload or {}).get('features'))
+        tech_stack = self._normalize_list((request_payload or {}).get('tech_stack'))
+        if not features:
+            features = ['core workflow based on free-form request']
+        return {
+            'name': name[:255],
+            'description': description[:255],
+            'features': features,
+            'tech_stack': tech_stack,
+        }
+
+    def _build_compact_context(self, context: dict) -> dict:
+        """Reduce interpreter context to the fields that help routing."""
+        projects = []
+        for project in context.get('projects', [])[:10]:
+            issues = []
+            for issue in project.get('open_issues', [])[:5]:
+                issues.append({'number': issue.get('number'), 'title': issue.get('title'), 'state': issue.get('state')})
+            projects.append(
+                {
+                    'project_id': project.get('project_id'),
+                    'name': project.get('name'),
+                    'description': project.get('description'),
+                    'repository': project.get('repository'),
+                    'open_pull_request': bool(project.get('open_pull_request')),
+                    'open_issues': issues,
+                }
+            )
+        return {
+            'chat_id': context.get('chat_id'),
+            'recent_chat_history': context.get('recent_chat_history', [])[:8],
+            'projects': projects,
+        }
+
+    def _normalize_routing(self, routing: dict | None, interpreted: dict, context: dict) -> dict:
+        """Normalize routing metadata returned by the LLM."""
+        routing = routing or {}
+        project_id = routing.get('project_id')
+        project_name = routing.get('project_name')
+        issue_number = routing.get('issue_number')
+        if issue_number in ('', None):
+            issue_number = None
+        elif isinstance(issue_number, str) and issue_number.isdigit():
+            issue_number = int(issue_number)
+        matched_project = None
+        for project in context.get('projects', []):
+            if project_id and project.get('project_id') == project_id:
+                matched_project = project
+                break
+            if project_name and project.get('name') == project_name:
+                matched_project = project
+                break
+        intent = str(routing.get('intent') or '').strip() or ('continue_project' if matched_project else 'new_project')
+        normalized = {
+            'intent': intent,
+            'project_id': matched_project.get('project_id') if matched_project else project_id,
+            'project_name': matched_project.get('name') if matched_project else (project_name or interpreted.get('name')),
+            'repo_name': routing.get('repo_name') if intent == 'new_project' else None,
+            'issue_number': issue_number,
+            'confidence': routing.get('confidence') or ('medium' if matched_project else 'low'),
+            'reasoning_summary': routing.get('reasoning_summary') or ('Matched prior project context' if matched_project else 'No strong prior project match found'),
+        }
+        if normalized['intent'] == 'new_project' and not normalized['repo_name']:
+            normalized['repo_name'] = self._derive_repo_name(normalized['project_name'] or interpreted.get('name') or 'Generated Project')
+        return normalized
+
+    def _normalize_list(self, value) -> list[str]:
+        if isinstance(value, list):
+            return [str(item).strip() for item in value if str(item).strip()]
+        if isinstance(value, str) and value.strip():
+            return [item.strip() for item in value.split(',') if item.strip()]
+        return []
+
+    def _derive_name(self, prompt_text: str) -> str:
+        """Derive a stable project name when the LLM does not provide one."""
+        first_line = prompt_text.splitlines()[0].strip()
+        quoted = re.search(r'["\']([^"\']{3,80})["\']', first_line)
+        if quoted:
+            return self._humanize_name(quoted.group(1))
+
+        noun_phrase = re.search(
+            r'(?:build|create|start|make|develop|generate|design|need|want)\s+'
+            r'(?:me\s+|us\s+|an?\s+|the\s+|new\s+|internal\s+|simple\s+|lightweight\s+|modern\s+|web\s+|mobile\s+)*'
+            r'([a-z0-9][a-z0-9\s-]{2,80}?(?:portal|dashboard|app|application|service|tool|system|platform|api|bot|assistant|website|site|workspace|tracker|manager))\b',
+            first_line,
+            flags=re.IGNORECASE,
+        )
+        if noun_phrase:
+            return self._humanize_name(noun_phrase.group(1))
+
+        cleaned = re.sub(r'[^A-Za-z0-9 ]+', ' ', first_line)
+        stopwords = {
+            'build', 'create', 'start', 'make', 'develop', 'generate', 'design', 'need', 'want', 'please', 'for', 'our', 'with', 'that', 'this',
+            'new', 'internal', 'simple', 'modern', 'web', 'mobile', 'app', 'application', 'tool', 'system',
+        }
+        tokens = [word for word in cleaned.split() if word and word.lower() not in stopwords]
+        if tokens:
+            return self._humanize_name(' '.join(tokens[:4]))
+        return 'Generated Project'
+
+    def _humanize_name(self, raw_name: str) -> str:
+        """Normalize a candidate project name into a readable title."""
+        cleaned = re.sub(r'[^A-Za-z0-9\s-]+', ' ', raw_name).strip(' -')
+        cleaned = re.sub(r'\s+', ' ', cleaned)
+        special_upper = {'api', 'crm', 'erp', 'cms', 'hr', 'it', 'ui', 'qa'}
+        words = []
+        for word in cleaned.split()[:6]:
+            lowered = word.lower()
+            words.append(lowered.upper() if lowered in special_upper else lowered.capitalize())
+        return ' '.join(words) or 'Generated Project'
+
+    def _derive_repo_name(self, project_name: str) -> str:
+        """Derive a repository slug from a human-readable project name."""
+        preferred = (project_name or 'project').strip().lower().replace(' ', '-')
+        sanitized = ''.join(ch if ch.isalnum() or ch in {'-', '_'} else '-' for ch in preferred)
+        while '--' in sanitized:
+            sanitized = sanitized.replace('--', '-')
+        return sanitized.strip('-') or 'project'
+
+    def _ensure_unique_repo_name(self, repo_name: str, reserved_names: set[str]) -> str:
+        """Choose a repository slug that does not collide with tracked or remote repositories."""
+        base_name = self._derive_repo_name(repo_name)
+        if base_name not in reserved_names:
+            return base_name
+        suffix = 2
+        while f'{base_name}-{suffix}' in reserved_names:
+            suffix += 1
+        return f'{base_name}-{suffix}'
+
+    def _normalize_project_identity(self, payload: dict, fallback_name: str) -> tuple[str, str]:
+        """Normalize model-proposed project and repository naming."""
+        project_name = self._humanize_name(str(payload.get('project_name') or payload.get('name') or fallback_name))
+        repo_name = self._derive_repo_name(str(payload.get('repo_name') or project_name))
+        return project_name, repo_name
+
+    def _heuristic_fallback(self, prompt_text: str, context: dict | None = None) -> tuple[dict, dict]:
+        """Fallback request extraction when Ollama is unavailable."""
+        lowered = prompt_text.lower()
+        tech_candidates = [
+            'python', 'fastapi', 'django', 'flask', 'postgresql', 'sqlite', 'react', 'vue', 'nicegui', 'docker'
+        ]
+        tech_stack = [candidate for candidate in tech_candidates if candidate in lowered]
+        sentences = [part.strip() for part in re.split(r'[\n\.]+', prompt_text) if part.strip()]
+        features = sentences[:3] or ['Implement the user request from free-form text']
+        interpreted = {
+            'name': self._derive_name(prompt_text),
+            'description': sentences[0][:255] if sentences else prompt_text[:255],
+            'features': features,
+            'tech_stack': tech_stack,
+        }
+        routing = self._heuristic_routing(prompt_text, context or {})
+        if routing.get('project_name'):
+            interpreted['name'] = routing['project_name']
+        return interpreted, routing
+
+    def _heuristic_routing(self, prompt_text: str, context: dict) -> dict:
+        """Best-effort routing when the LLM is unavailable."""
+        lowered = prompt_text.lower()
+        explicit_new = any(token in lowered for token in ['new project', 'start a new project', 'create a new project', 'build a new app'])
+        referenced_issue = self._extract_issue_number(prompt_text)
+        recent_history = context.get('recent_chat_history', [])
+        projects = context.get('projects', [])
+        last_project_id = recent_history[0].get('project_id') if recent_history else None
+        last_issue = ((recent_history[0].get('related_issue') or {}).get('number') if recent_history else None)
+
+        matched_project = None
+        for project in projects:
+            name = (project.get('name') or '').lower()
+            repo = ((project.get('repository') or {}).get('name') or '').lower()
+            if name and name in lowered:
+                matched_project = project
+                break
+            if repo and repo in lowered:
+                matched_project = project
+                break
+        if matched_project is None and not explicit_new:
+            follow_up_tokens = ['also', 'continue', 'for this project', 'for that project', 'work on this', 'work on that', 'fix that', 'add this']
+            if any(token in lowered for token in follow_up_tokens) and last_project_id:
+                matched_project = next((project for project in projects if project.get('project_id') == last_project_id), None)
+        issue_number = referenced_issue
+        if issue_number is None and any(token in lowered for token in ['that issue', 'this issue', 'the issue']) and last_issue is not None:
+            issue_number = last_issue
+        intent = 'new_project' if explicit_new or matched_project is None else 'continue_project'
+        return {
+            'intent': intent,
+            'project_id': matched_project.get('project_id') if matched_project else None,
+            'project_name': matched_project.get('name') if matched_project else self._derive_name(prompt_text),
+            'repo_name': None if matched_project else self._derive_repo_name(self._derive_name(prompt_text)),
+            'issue_number': issue_number,
+            'confidence': 'medium' if matched_project or explicit_new else 'low',
+            'reasoning_summary': 'Heuristic routing from chat history and project names.',
+        }
+
+    def _extract_issue_number(self, prompt_text: str) -> int | None:
+        match = re.search(r'(?:#|issue\s+)(\d+)', prompt_text, flags=re.IGNORECASE)
+        return int(match.group(1)) if match else None

ai_software_factory/agents/telegram.py

@@ -1,8 +1,6 @@
 """Telegram bot integration for n8n webhook."""
 
 import asyncio
-import json
-import re
 from typing import Optional
 
 
@@ -12,6 +10,59 @@ class TelegramHandler:
     def __init__(self, webhook_url: str):
         self.webhook_url = webhook_url
         self.api_url = "https://api.telegram.org/bot"
+
+    def build_prompt_guide_message(self, backend_url: str | None = None) -> str:
+        """Build a Telegram message explaining the expected prompt format."""
+        lines = [
+            "AI Software Factory is listening in this chat.",
+            "",
+            "You can send free-form software requests in normal language.",
+            "",
+            "Example:",
+            "Build an internal inventory portal for our warehouse team.",
+            "It should support role-based login, stock dashboards, and purchase orders.",
+            "Prefer FastAPI, PostgreSQL, and a simple web UI.",
+            "",
+            "The backend will interpret the request and turn it into a structured project plan.",
+        ]
+        if backend_url:
+            lines.extend(["", f"Backend target: {backend_url}"])
+        return "\n".join(lines)
+
+    async def send_message(self, bot_token: str, chat_id: str | int, text: str) -> dict:
+        """Send a direct Telegram message using the configured bot."""
+        if not bot_token:
+            return {"status": "error", "message": "Telegram bot token is not configured"}
+        if chat_id in (None, ""):
+            return {"status": "error", "message": "Telegram chat id is not configured"}
+
+        api_endpoint = f"{self.api_url}{bot_token}/sendMessage"
+
+        try:
+            import aiohttp
+            async with aiohttp.ClientSession() as session:
+                async with session.post(
+                    api_endpoint,
+                    json={
+                        "chat_id": str(chat_id),
+                        "text": text,
+                    },
+                ) as resp:
+                    payload = await resp.json()
+                    if 200 <= resp.status < 300 and payload.get("ok"):
+                        return {
+                            "status": "success",
+                            "message": "Telegram prompt guide sent successfully",
+                            "payload": payload,
+                        }
+                    description = payload.get("description") or payload.get("message") or str(payload)
+                    return {
+                        "status": "error",
+                        "message": f"Telegram API returned {resp.status}: {description}",
+                        "payload": payload,
+                    }
+        except Exception as exc:
+            return {"status": "error", "message": str(exc)}
     
     async def handle_message(self, message_data: dict) -> dict:
         """Handle incoming Telegram message."""

ai_software_factory/config.py

@@ -1,5 +1,6 @@
 """Configuration settings for AI Software Factory."""
 
+import json
 import os
 from typing import Optional
 from pathlib import Path
@@ -24,6 +25,34 @@ class Settings(BaseSettings):
     # Ollama settings computed from environment
     OLLAMA_URL: str = "http://ollama:11434"
     OLLAMA_MODEL: str = "llama3"
+    LLM_GUARDRAIL_PROMPT: str = (
+        "You are operating inside AI Software Factory. Follow the requested schema exactly, "
+        "treat provided tool outputs as authoritative, and do not invent repositories, issues, pull requests, or delivery facts."
+    )
+    LLM_REQUEST_INTERPRETER_GUARDRAIL_PROMPT: str = (
+        "For routing and request interpretation: never select archived projects, prefer tracked project IDs from tool outputs, and only reference issues that are explicit in the prompt or available tool data."
+    )
+    LLM_CHANGE_SUMMARY_GUARDRAIL_PROMPT: str = (
+        "For summaries: only describe facts present in the provided context and tool outputs. Never claim a repository, commit, or pull request exists unless it is present in the supplied data."
+    )
+    LLM_PROJECT_NAMING_GUARDRAIL_PROMPT: str = (
+        "For project naming: prefer clear, product-like names and repository slugs that match the user's intent. Avoid reusing tracked project identities unless the request is clearly asking for an existing project."
+    )
+    LLM_PROJECT_NAMING_SYSTEM_PROMPT: str = (
+        "You name newly requested software projects. Return only JSON with keys project_name, repo_name, and rationale. Project names should be concise human-readable titles. Repo names should be lowercase kebab-case slugs suitable for a Gitea repository name."
+    )
+    LLM_PROJECT_ID_GUARDRAIL_PROMPT: str = (
+        "For project ids: produce short stable slugs for newly created projects. Avoid collisions with known project ids and keep ids lowercase with hyphens."
+    )
+    LLM_PROJECT_ID_SYSTEM_PROMPT: str = (
+        "You derive stable project ids for new projects. Return only JSON with keys project_id and rationale. project_id must be a short lowercase kebab-case slug without spaces."
+    )
+    LLM_TOOL_ALLOWLIST: str = "gitea_project_catalog,gitea_project_state,gitea_project_issues,gitea_pull_requests"
+    LLM_TOOL_CONTEXT_LIMIT: int = 5
+    LLM_LIVE_TOOL_ALLOWLIST: str = "gitea_lookup_issue,gitea_lookup_pull_request"
+    LLM_LIVE_TOOL_STAGE_ALLOWLIST: str = "request_interpretation,change_summary"
+    LLM_LIVE_TOOL_STAGE_TOOL_MAP: str = ""
+    LLM_MAX_TOOL_CALL_ROUNDS: int = 1
 
     # Gitea settings
     GITEA_URL: str = "https://gitea.yourserver.com"
@@ -131,6 +160,97 @@ class Settings(BaseSettings):
         """Get Ollama URL with trimmed whitespace."""
         return self.OLLAMA_URL.strip()
 
+    @property
+    def llm_guardrail_prompt(self) -> str:
+        """Get the global guardrail prompt used for all external LLM calls."""
+        return self.LLM_GUARDRAIL_PROMPT.strip()
+
+    @property
+    def llm_request_interpreter_guardrail_prompt(self) -> str:
+        """Get the request-interpretation specific guardrail prompt."""
+        return self.LLM_REQUEST_INTERPRETER_GUARDRAIL_PROMPT.strip()
+
+    @property
+    def llm_change_summary_guardrail_prompt(self) -> str:
+        """Get the change-summary specific guardrail prompt."""
+        return self.LLM_CHANGE_SUMMARY_GUARDRAIL_PROMPT.strip()
+
+    @property
+    def llm_project_naming_guardrail_prompt(self) -> str:
+        """Get the project-naming specific guardrail prompt."""
+        return self.LLM_PROJECT_NAMING_GUARDRAIL_PROMPT.strip()
+
+    @property
+    def llm_project_naming_system_prompt(self) -> str:
+        """Get the project-naming system prompt."""
+        return self.LLM_PROJECT_NAMING_SYSTEM_PROMPT.strip()
+
+    @property
+    def llm_project_id_guardrail_prompt(self) -> str:
+        """Get the project-id naming specific guardrail prompt."""
+        return self.LLM_PROJECT_ID_GUARDRAIL_PROMPT.strip()
+
+    @property
+    def llm_project_id_system_prompt(self) -> str:
+        """Get the project-id naming system prompt."""
+        return self.LLM_PROJECT_ID_SYSTEM_PROMPT.strip()
+
+    @property
+    def llm_tool_allowlist(self) -> list[str]:
+        """Get the allowed LLM tool names as a normalized list."""
+        return [item.strip() for item in self.LLM_TOOL_ALLOWLIST.split(',') if item.strip()]
+
+    @property
+    def llm_tool_context_limit(self) -> int:
+        """Get the number of items to expose per mediated tool payload."""
+        return max(int(self.LLM_TOOL_CONTEXT_LIMIT), 1)
+
+    @property
+    def llm_live_tool_allowlist(self) -> list[str]:
+        """Get the allowed live tool-call names for model-driven lookup requests."""
+        return [item.strip() for item in self.LLM_LIVE_TOOL_ALLOWLIST.split(',') if item.strip()]
+
+    @property
+    def llm_live_tool_stage_allowlist(self) -> list[str]:
+        """Get the LLM stages where live tool requests are enabled."""
+        return [item.strip() for item in self.LLM_LIVE_TOOL_STAGE_ALLOWLIST.split(',') if item.strip()]
+
+    @property
+    def llm_live_tool_stage_tool_map(self) -> dict[str, list[str]]:
+        """Get an optional per-stage live tool map that overrides the simple stage allowlist."""
+        raw = (self.LLM_LIVE_TOOL_STAGE_TOOL_MAP or '').strip()
+        if not raw:
+            return {}
+        try:
+            parsed = json.loads(raw)
+        except Exception:
+            return {}
+        if not isinstance(parsed, dict):
+            return {}
+        allowed_tools = set(self.llm_live_tool_allowlist)
+        normalized: dict[str, list[str]] = {}
+        for stage, tools in parsed.items():
+            if not isinstance(stage, str):
+                continue
+            if not isinstance(tools, list):
+                continue
+            normalized[stage.strip()] = [str(tool).strip() for tool in tools if str(tool).strip() in allowed_tools]
+        return normalized
+
+    def llm_live_tools_for_stage(self, stage: str) -> list[str]:
+        """Return live tools enabled for a specific LLM stage."""
+        stage_map = self.llm_live_tool_stage_tool_map
+        if stage_map:
+            return stage_map.get(stage, [])
+        if stage not in set(self.llm_live_tool_stage_allowlist):
+            return []
+        return self.llm_live_tool_allowlist
+
+    @property
+    def llm_max_tool_call_rounds(self) -> int:
+        """Get the maximum number of model-driven live tool-call rounds per LLM request."""
+        return max(int(self.LLM_MAX_TOOL_CALL_ROUNDS), 0)
+
     @property
     def gitea_url(self) -> str:
         """Get Gitea URL with trimmed whitespace."""

ai_software_factory/frontend.py

@@ -24,6 +24,7 @@ def init(fastapi_app: FastAPI, storage_secret: str = 'Secr2t!') -> None:
     """
     
     def render_dashboard_page() -> None:
+        ui.page_title('AI Software Factory')
         create_dashboard()
 
         # NOTE dark mode will be persistent for each user across tabs and server restarts

ai_software_factory/main.py

@@ -27,17 +27,25 @@ from sqlalchemy.orm import Session
 try:
     from . import __version__, frontend
     from . import database as database_module
+    from .agents.change_summary import ChangeSummaryGenerator
     from .agents.database_manager import DatabaseManager
+    from .agents.request_interpreter import RequestInterpreter
+    from .agents.llm_service import LLMServiceClient
     from .agents.orchestrator import AgentOrchestrator
     from .agents.n8n_setup import N8NSetupAgent
+    from .agents.prompt_workflow import PromptWorkflowManager
     from .agents.ui_manager import UIManager
     from .models import ProjectHistory, ProjectLog, SystemLog
 except ImportError:
     import frontend
     import database as database_module
+    from agents.change_summary import ChangeSummaryGenerator
     from agents.database_manager import DatabaseManager
+    from agents.request_interpreter import RequestInterpreter
+    from agents.llm_service import LLMServiceClient
     from agents.orchestrator import AgentOrchestrator
     from agents.n8n_setup import N8NSetupAgent
+    from agents.prompt_workflow import PromptWorkflowManager
     from agents.ui_manager import UIManager
     from models import ProjectHistory, ProjectLog, SystemLog
 
@@ -79,12 +87,99 @@ class N8NSetupRequest(BaseModel):
     force_update: bool = False
 
 
+class FreeformSoftwareRequest(BaseModel):
+    """Request body for free-form software generation."""
+
+    prompt_text: str = Field(min_length=1)
+    source: str = 'telegram'
+    chat_id: str | None = None
+    chat_type: str | None = None
+
+
+class GiteaRepositoryOnboardRequest(BaseModel):
+    """Request body for onboarding a manually created Gitea repository."""
+
+    repo_name: str = Field(min_length=1, max_length=255)
+    owner: str | None = None
+    sync_commits: bool = True
+    commit_limit: int = Field(default=25, ge=1, le=200)
+
+
 def _build_project_id(name: str) -> str:
     """Create a stable project id from the requested name."""
     slug = PROJECT_ID_PATTERN.sub("-", name.strip().lower()).strip("-") or "project"
     return f"{slug}-{uuid4().hex[:8]}"
 
 
+def _build_project_slug(name: str) -> str:
+    """Normalize a project name into a kebab-case identifier slug."""
+    return PROJECT_ID_PATTERN.sub("-", name.strip().lower()).strip("-") or "project"
+
+
+def _ensure_unique_identifier(base_slug: str, reserved_ids: set[str]) -> str:
+    """Return a unique identifier using deterministic numeric suffixes when needed."""
+    normalized = _build_project_slug(base_slug)
+    if normalized not in reserved_ids:
+        return normalized
+    suffix = 2
+    while f"{normalized}-{suffix}" in reserved_ids:
+        suffix += 1
+    return f"{normalized}-{suffix}"
+
+
+def _build_project_identity_context(manager: DatabaseManager) -> list[dict]:
+    """Build a compact project catalog for naming stages."""
+    projects = []
+    for history in manager.get_all_projects(include_archived=True):
+        repository = manager._get_project_repository(history) or {}
+        projects.append(
+            {
+                'project_id': history.project_id,
+                'name': history.project_name,
+                'description': history.description,
+                'repository': {
+                    'owner': repository.get('owner'),
+                    'name': repository.get('name'),
+                },
+            }
+        )
+    return projects
+
+
+async def _derive_project_id_for_request(
+    request: SoftwareRequest,
+    *,
+    prompt_text: str,
+    prompt_routing: dict | None,
+    existing_projects: list[dict],
+) -> tuple[str, dict | None]:
+    """Derive a stable project id for a newly created project."""
+    reserved_ids = {str(project.get('project_id')).strip() for project in existing_projects if project.get('project_id')}
+    fallback_id = _ensure_unique_identifier((prompt_routing or {}).get('project_name') or request.name, reserved_ids)
+    user_prompt = (
+        f"Original user prompt:\n{prompt_text}\n\n"
+        f"Structured request:\n{json.dumps({'name': request.name, 'description': request.description, 'features': request.features, 'tech_stack': request.tech_stack}, indent=2)}\n\n"
+        f"Naming context:\n{json.dumps(prompt_routing or {}, indent=2)}\n\n"
+        f"Reserved project ids:\n{json.dumps(sorted(reserved_ids))}\n\n"
+        "Suggest the best stable project id for this new project."
+    )
+    content, trace = await LLMServiceClient().chat_with_trace(
+        stage='project_id_naming',
+        system_prompt=database_module.settings.llm_project_id_system_prompt,
+        user_prompt=user_prompt,
+        tool_context_input={'projects': existing_projects},
+        expect_json=True,
+    )
+    if content:
+        try:
+            parsed = json.loads(content)
+            candidate = parsed.get('project_id') or parsed.get('slug') or request.name
+            return _ensure_unique_identifier(str(candidate), reserved_ids), trace
+        except Exception:
+            pass
+    return fallback_id, trace
+
+
 def _serialize_project(history: ProjectHistory) -> dict:
     """Serialize a project history row for API responses."""
     return {
@@ -144,11 +239,164 @@ def _compose_prompt_text(request: SoftwareRequest) -> str:
     )
 
 
+async def _run_generation(
+    request: SoftwareRequest,
+    db: Session,
+    prompt_text: str | None = None,
+    prompt_actor: str = 'api',
+    prompt_source_context: dict | None = None,
+    prompt_routing: dict | None = None,
+    preferred_project_id: str | None = None,
+    repo_name_override: str | None = None,
+    related_issue: dict | None = None,
+) -> dict:
+    """Run the shared generation pipeline for a structured request."""
+    database_module.init_db()
+
+    manager = DatabaseManager(db)
+    is_explicit_new_project = (prompt_routing or {}).get('intent') == 'new_project'
+    reusable_history = manager.get_project_by_id(preferred_project_id, include_archived=False) if preferred_project_id else (None if is_explicit_new_project else manager.get_latest_project_by_name(request.name))
+    if reusable_history and database_module.settings.gitea_url and database_module.settings.gitea_token:
+        try:
+            from .agents.gitea import GiteaAPI
+        except ImportError:
+            from agents.gitea import GiteaAPI
+        manager.sync_pull_request_states(
+            GiteaAPI(
+                token=database_module.settings.GITEA_TOKEN,
+                base_url=database_module.settings.GITEA_URL,
+                owner=database_module.settings.GITEA_OWNER,
+                repo=database_module.settings.GITEA_REPO or '',
+            ),
+            project_id=reusable_history.project_id,
+        )
+    project_id_trace = None
+    resolved_prompt_text = prompt_text or _compose_prompt_text(request)
+    if preferred_project_id and reusable_history is not None:
+        project_id = reusable_history.project_id
+    elif reusable_history and not is_explicit_new_project and manager.get_open_pull_request(project_id=reusable_history.project_id):
+        project_id = reusable_history.project_id
+    else:
+        if is_explicit_new_project or prompt_text:
+            project_id, project_id_trace = await _derive_project_id_for_request(
+                request,
+                prompt_text=resolved_prompt_text,
+                prompt_routing=prompt_routing,
+                existing_projects=_build_project_identity_context(manager),
+            )
+        else:
+            project_id = _build_project_id(request.name)
+        reusable_history = None
+    orchestrator = AgentOrchestrator(
+        project_id=project_id,
+        project_name=request.name,
+        description=request.description,
+        features=request.features,
+        tech_stack=request.tech_stack,
+        db=db,
+        prompt_text=resolved_prompt_text,
+        prompt_actor=prompt_actor,
+        existing_history=reusable_history,
+        prompt_source_context=prompt_source_context,
+        prompt_routing=prompt_routing,
+        repo_name_override=repo_name_override,
+        related_issue_hint=related_issue,
+    )
+    result = await orchestrator.run()
+
+    manager = DatabaseManager(db)
+    manager.log_system_event(
+        component='api',
+        level='INFO' if result['status'] == 'completed' else 'ERROR',
+        message=f"Generated project {project_id} with {len(result.get('changed_files', []))} artifact(s)",
+    )
+
+    history = manager.get_project_by_id(project_id)
+    project_logs = manager.get_project_logs(history.id)
+    response_data = _serialize_project(history)
+    response_data['logs'] = [_serialize_project_log(log) for log in project_logs]
+    response_data['ui_data'] = result.get('ui_data')
+    response_data['features'] = request.features
+    response_data['tech_stack'] = request.tech_stack
+    response_data['project_root'] = result.get('project_root', str(_project_root(project_id)))
+    response_data['changed_files'] = result.get('changed_files', [])
+    response_data['repository'] = result.get('repository')
+    response_data['related_issue'] = result.get('related_issue') or (result.get('ui_data') or {}).get('related_issue')
+    response_data['pull_request'] = result.get('pull_request') or manager.get_open_pull_request(project_id=project_id)
+    if project_id_trace:
+        manager.log_llm_trace(
+            project_id=project_id,
+            history_id=history.id if history else None,
+            prompt_id=orchestrator.prompt_audit.id if orchestrator.prompt_audit else None,
+            stage=project_id_trace['stage'],
+            provider=project_id_trace['provider'],
+            model=project_id_trace['model'],
+            system_prompt=project_id_trace['system_prompt'],
+            user_prompt=project_id_trace['user_prompt'],
+            assistant_response=project_id_trace['assistant_response'],
+            raw_response=project_id_trace.get('raw_response'),
+            fallback_used=project_id_trace.get('fallback_used', False),
+        )
+    summary_context = {
+        'name': response_data['name'],
+        'description': response_data['description'],
+        'features': response_data['features'],
+        'tech_stack': response_data['tech_stack'],
+        'changed_files': response_data['changed_files'],
+        'repository_url': (
+            (response_data.get('repository') or {}).get('url')
+            if isinstance(response_data.get('repository'), dict)
+            and (response_data.get('repository') or {}).get('status') in {'created', 'exists', 'ready', 'shared'}
+            else None
+        ),
+        'repository_status': (response_data.get('repository') or {}).get('status') if isinstance(response_data.get('repository'), dict) else None,
+        'pull_request_url': (response_data.get('pull_request') or {}).get('pr_url') if isinstance(response_data.get('pull_request'), dict) else None,
+        'pull_request_state': (response_data.get('pull_request') or {}).get('pr_state') if isinstance(response_data.get('pull_request'), dict) else None,
+        'related_issue': response_data.get('related_issue'),
+        'message': response_data.get('message'),
+        'logs': [log.get('message', '') for log in response_data.get('logs', []) if isinstance(log, dict)],
+    }
+    summary_message, summary_trace = await ChangeSummaryGenerator().summarize_with_trace(summary_context)
+    if orchestrator.db_manager and orchestrator.history and orchestrator.prompt_audit:
+        orchestrator.db_manager.log_llm_trace(
+            project_id=project_id,
+            history_id=orchestrator.history.id,
+            prompt_id=orchestrator.prompt_audit.id,
+            stage=summary_trace['stage'],
+            provider=summary_trace['provider'],
+            model=summary_trace['model'],
+            system_prompt=summary_trace['system_prompt'],
+            user_prompt=summary_trace['user_prompt'],
+            assistant_response=summary_trace['assistant_response'],
+            raw_response=summary_trace.get('raw_response'),
+            fallback_used=summary_trace.get('fallback_used', False),
+        )
+    response_data['summary_message'] = summary_message
+    response_data['pull_request'] = result.get('pull_request') or manager.get_open_pull_request(project_id=project_id)
+    return {'status': result['status'], 'data': response_data, 'summary_message': summary_message}
+
+
 def _project_root(project_id: str) -> Path:
     """Resolve the filesystem location for a generated project."""
     return database_module.settings.projects_root / project_id
 
 
+def _create_gitea_api():
+    """Create a configured Gitea client or raise an HTTP error if unavailable."""
+    if not database_module.settings.gitea_url or not database_module.settings.gitea_token:
+        raise HTTPException(status_code=400, detail='Gitea integration is not configured')
+    try:
+        from .agents.gitea import GiteaAPI
+    except ImportError:
+        from agents.gitea import GiteaAPI
+    return GiteaAPI(
+        token=database_module.settings.GITEA_TOKEN,
+        base_url=database_module.settings.GITEA_URL,
+        owner=database_module.settings.GITEA_OWNER,
+        repo=database_module.settings.GITEA_REPO or '',
+    )
+
+
 def _resolve_n8n_api_url(explicit_url: str | None = None) -> str:
     """Resolve the effective n8n API URL from explicit input or settings."""
     if explicit_url and explicit_url.strip():
@@ -171,7 +419,9 @@ def read_api_info():
             '/',
             '/api',
             '/health',
+            '/llm/runtime',
             '/generate',
+            '/generate/text',
             '/projects',
             '/status/{project_id}',
             '/audit/projects',
@@ -179,8 +429,20 @@ def read_api_info():
             '/audit/system/logs',
             '/audit/prompts',
             '/audit/changes',
+            '/audit/issues',
+            '/audit/commit-context',
+            '/audit/timeline',
+            '/audit/llm-traces',
+            '/audit/pull-requests',
             '/audit/lineage',
             '/audit/correlations',
+            '/projects/{project_id}/archive',
+            '/projects/{project_id}/unarchive',
+            '/projects/{project_id}',
+            '/projects/{project_id}/prompts/{prompt_id}/undo',
+            '/projects/{project_id}/sync-repository',
+            '/gitea/repos',
+            '/gitea/repos/onboard',
             '/n8n/health',
             '/n8n/setup',
         ],
@@ -199,50 +461,116 @@ def health_check():
     }
 
 
+@app.get('/llm/runtime')
+def get_llm_runtime():
+    """Return the active external LLM runtime, guardrail, and tool configuration."""
+    return LLMServiceClient().get_runtime_configuration()
+
+
 @app.post('/generate')
 async def generate_software(request: SoftwareRequest, db: DbSession):
     """Create and record a software-generation request."""
-    database_module.init_db()
+    return await _run_generation(request, db)
 
-    project_id = _build_project_id(request.name)
-    prompt_text = _compose_prompt_text(request)
-    orchestrator = AgentOrchestrator(
-        project_id=project_id,
-        project_name=request.name,
-        description=request.description,
-        features=request.features,
-        tech_stack=request.tech_stack,
-        db=db,
-        prompt_text=prompt_text,
-    )
-    result = await orchestrator.run()
+
+@app.post('/generate/text')
+async def generate_software_from_text(request: FreeformSoftwareRequest, db: DbSession):
+    """Interpret a free-form request and run generation."""
+    if (
+        request.source == 'telegram'
+        and database_module.settings.telegram_chat_id
+        and request.chat_id
+        and str(request.chat_id) != str(database_module.settings.telegram_chat_id)
+    ):
+        return {
+            'status': 'ignored',
+            'message': f"Ignoring Telegram message from chat {request.chat_id}",
+            'source': {
+                'type': request.source,
+                'chat_id': request.chat_id,
+                'chat_type': request.chat_type,
+            },
+        }
 
     manager = DatabaseManager(db)
-    manager.log_system_event(
-        component='api',
-        level='INFO' if result['status'] == 'completed' else 'ERROR',
-        message=f"Generated project {project_id} with {len(result.get('changed_files', []))} artifact(s)",
+    interpreter_context = manager.get_interpreter_context(chat_id=request.chat_id, source=request.source)
+    interpreted, interpretation_trace = await RequestInterpreter().interpret_with_trace(
+        request.prompt_text,
+        context=interpreter_context,
     )
-
-    history = manager.get_project_by_id(project_id)
-    project_logs = manager.get_project_logs(history.id)
-    response_data = _serialize_project(history)
-    response_data['logs'] = [_serialize_project_log(log) for log in project_logs]
-    response_data['ui_data'] = result.get('ui_data')
-    response_data['features'] = request.features
-    response_data['tech_stack'] = request.tech_stack
-    response_data['project_root'] = result.get('project_root', str(_project_root(project_id)))
-    response_data['changed_files'] = result.get('changed_files', [])
-    response_data['repository'] = result.get('repository')
-
-    return {'status': result['status'], 'data': response_data}
+    routing = interpretation_trace.get('routing') or {}
+    selected_history = manager.get_project_by_id(routing.get('project_id'), include_archived=False) if routing.get('project_id') else None
+    if selected_history is not None and routing.get('intent') != 'new_project':
+        interpreted['name'] = selected_history.project_name
+        interpreted['description'] = selected_history.description or interpreted['description']
+    structured_request = SoftwareRequest(**interpreted)
+    response = await _run_generation(
+        structured_request,
+        db,
+        prompt_text=request.prompt_text,
+        prompt_actor=request.source,
+        prompt_source_context={
+            'chat_id': request.chat_id,
+            'chat_type': request.chat_type,
+        },
+        prompt_routing=routing,
+        preferred_project_id=routing.get('project_id') if routing.get('intent') != 'new_project' else None,
+        repo_name_override=routing.get('repo_name') if routing.get('intent') == 'new_project' else None,
+        related_issue={'number': routing.get('issue_number')} if routing.get('issue_number') is not None else None,
+    )
+    project_data = response.get('data', {})
+    if project_data.get('history_id') is not None:
+        manager = DatabaseManager(db)
+        prompts = manager.get_prompt_events(project_id=project_data.get('project_id'))
+        prompt_id = prompts[0]['id'] if prompts else None
+        manager.log_llm_trace(
+            project_id=project_data.get('project_id'),
+            history_id=project_data.get('history_id'),
+            prompt_id=prompt_id,
+            stage=interpretation_trace['stage'],
+            provider=interpretation_trace['provider'],
+            model=interpretation_trace['model'],
+            system_prompt=interpretation_trace['system_prompt'],
+            user_prompt=interpretation_trace['user_prompt'],
+            assistant_response=interpretation_trace['assistant_response'],
+            raw_response=interpretation_trace.get('raw_response'),
+            fallback_used=interpretation_trace.get('fallback_used', False),
+        )
+        naming_trace = interpretation_trace.get('project_naming')
+        if naming_trace:
+            manager.log_llm_trace(
+                project_id=project_data.get('project_id'),
+                history_id=project_data.get('history_id'),
+                prompt_id=prompt_id,
+                stage=naming_trace['stage'],
+                provider=naming_trace['provider'],
+                model=naming_trace['model'],
+                system_prompt=naming_trace['system_prompt'],
+                user_prompt=naming_trace['user_prompt'],
+                assistant_response=naming_trace['assistant_response'],
+                raw_response=naming_trace.get('raw_response'),
+                fallback_used=naming_trace.get('fallback_used', False),
+            )
+    response['interpreted_request'] = interpreted
+    response['routing'] = routing
+    response['llm_trace'] = interpretation_trace
+    response['source'] = {
+        'type': request.source,
+        'chat_id': request.chat_id,
+        'chat_type': request.chat_type,
+    }
+    return response
 
 
 @app.get('/projects')
-def list_projects(db: DbSession):
+def list_projects(
+    db: DbSession,
+    include_archived: bool = Query(default=False),
+    archived_only: bool = Query(default=False),
+):
     """List recorded projects."""
     manager = DatabaseManager(db)
-    projects = manager.get_all_projects()
+    projects = manager.get_all_projects(include_archived=include_archived, archived_only=archived_only)
     return {'projects': [_serialize_project(project) for project in projects]}
 
 
@@ -285,6 +613,68 @@ def get_code_change_audit(db: DbSession, project_id: str | None = Query(default=
     return {'changes': [_serialize_audit_item(item) for item in manager.get_code_changes(project_id=project_id)]}
 
 
+@app.get('/audit/issues')
+def get_issue_audit(
+    db: DbSession,
+    project_id: str | None = Query(default=None),
+    state: str | None = Query(default=None),
+):
+    """Return tracked repository issues and issue-work events."""
+    manager = DatabaseManager(db)
+    return {
+        'issues': manager.get_repository_issues(project_id=project_id, state=state),
+        'issue_work': manager.get_issue_work_events(project_id=project_id),
+    }
+
+
+@app.get('/audit/commit-context')
+def get_commit_context_audit(
+    db: DbSession,
+    commit_hash: str = Query(min_length=4),
+    project_id: str | None = Query(default=None),
+    branch_scope: str | None = Query(default=None, pattern='^(main|pr|manual)?$'),
+):
+    """Return the recorded context explaining how a commit came to be."""
+    manager = DatabaseManager(db)
+    context = manager.get_commit_context(commit_hash=commit_hash, project_id=project_id, branch_scope=branch_scope)
+    if context is None:
+        raise HTTPException(status_code=404, detail='Commit context not found')
+    return context
+
+
+@app.get('/audit/timeline')
+def get_project_timeline_audit(
+    db: DbSession,
+    project_id: str = Query(min_length=1),
+    branch_scope: str | None = Query(default=None, pattern='^(main|pr|manual)?$'),
+):
+    """Return the mixed audit timeline for one project."""
+    manager = DatabaseManager(db)
+    return {'timeline': manager.get_project_timeline(project_id=project_id, branch_scope=branch_scope)}
+
+
+@app.get('/audit/llm-traces')
+def get_llm_trace_audit(
+    db: DbSession,
+    project_id: str | None = Query(default=None),
+    prompt_id: int | None = Query(default=None),
+    stage: str | None = Query(default=None),
+    model: str | None = Query(default=None),
+    search: str | None = Query(default=None),
+):
+    """Return persisted LLM traces."""
+    manager = DatabaseManager(db)
+    return {
+        'llm_traces': manager.get_llm_traces(
+            project_id=project_id,
+            prompt_id=prompt_id,
+            stage=stage,
+            model=model,
+            search_query=search,
+        )
+    }
+
+
 @app.get('/audit/lineage')
 def get_prompt_change_lineage(db: DbSession, project_id: str | None = Query(default=None)):
     """Return explicit prompt-to-code lineage rows."""
@@ -299,6 +689,141 @@ def get_prompt_change_correlations(db: DbSession, project_id: str | None = Query
     return {'correlations': manager.get_prompt_change_correlations(project_id=project_id)}
 
 
+@app.get('/audit/pull-requests')
+def get_pull_request_audit(db: DbSession, project_id: str | None = Query(default=None), open_only: bool = Query(default=False)):
+    """Return tracked pull requests for generated projects."""
+    manager = DatabaseManager(db)
+    return {'pull_requests': manager.get_pull_requests(project_id=project_id, only_open=open_only)}
+
+
+@app.post('/projects/{project_id}/prompts/{prompt_id}/undo')
+async def undo_prompt_changes(project_id: str, prompt_id: int, db: DbSession):
+    """Undo all changes associated with a specific prompt."""
+    manager = DatabaseManager(db)
+    history = manager.get_project_by_id(project_id)
+    if history is None:
+        raise HTTPException(status_code=404, detail='Project not found')
+    if history.status == 'archived':
+        raise HTTPException(status_code=400, detail='Archived projects cannot be modified')
+    result = await PromptWorkflowManager(db).undo_prompt(project_id=project_id, prompt_id=prompt_id)
+    if result.get('status') == 'error':
+        raise HTTPException(status_code=400, detail=result.get('message', 'Undo failed'))
+    return result
+
+
+@app.post('/projects/{project_id}/archive')
+def archive_project(project_id: str, db: DbSession):
+    """Archive a project so it no longer participates in active automation."""
+    manager = DatabaseManager(db)
+    result = manager.archive_project(project_id)
+    if result.get('status') == 'error':
+        raise HTTPException(status_code=404, detail=result.get('message', 'Archive failed'))
+    return result
+
+
+@app.post('/projects/{project_id}/unarchive')
+def unarchive_project(project_id: str, db: DbSession):
+    """Restore an archived project back into the active automation set."""
+    manager = DatabaseManager(db)
+    result = manager.unarchive_project(project_id)
+    if result.get('status') == 'error':
+        raise HTTPException(status_code=404, detail=result.get('message', 'Restore failed'))
+    return result
+
+
+@app.delete('/projects/{project_id}')
+def delete_project(project_id: str, db: DbSession):
+    """Delete a project, its local project directory, and project-scoped DB traces."""
+    manager = DatabaseManager(db)
+    audit_data = manager.get_project_audit_data(project_id)
+    if audit_data.get('project') is None:
+        raise HTTPException(status_code=404, detail='Project not found')
+
+    repository = audit_data.get('repository') or audit_data['project'].get('repository') or {}
+    remote_delete = None
+    if repository and repository.get('mode') != 'shared' and repository.get('owner') and repository.get('name') and database_module.settings.gitea_url and database_module.settings.gitea_token:
+        remote_delete = _create_gitea_api().delete_repo_sync(owner=repository.get('owner'), repo=repository.get('name'))
+        if remote_delete.get('error') and remote_delete.get('status_code') not in {404, None}:
+            raise HTTPException(status_code=502, detail=remote_delete.get('error'))
+
+    result = manager.delete_project(project_id)
+    if result.get('status') == 'error':
+        raise HTTPException(status_code=400, detail=result.get('message', 'Project deletion failed'))
+    result['remote_repository_deleted'] = bool(remote_delete and not remote_delete.get('error'))
+    result['remote_repository'] = repository if repository else None
+    return result
+
+
+@app.post('/projects/{project_id}/sync-repository')
+def sync_project_repository(project_id: str, db: DbSession, commit_limit: int = Query(default=25, ge=1, le=200)):
+    """Import recent repository activity from Gitea for a tracked project."""
+    manager = DatabaseManager(db)
+    history = manager.get_project_by_id(project_id)
+    if history is None:
+        raise HTTPException(status_code=404, detail='Project not found')
+    if history.status == 'archived':
+        raise HTTPException(status_code=400, detail='Archived projects cannot be synced')
+    gitea_api = _create_gitea_api()
+    result = manager.sync_repository_activity(project_id=project_id, gitea_api=gitea_api, commit_limit=commit_limit)
+    if result.get('status') == 'error':
+        raise HTTPException(status_code=400, detail=result.get('message', 'Repository sync failed'))
+    manager.sync_repository_issues(project_id=project_id, gitea_api=gitea_api, state='open')
+    return result
+
+
+@app.get('/gitea/repos')
+def list_gitea_repositories(db: DbSession, owner: str | None = Query(default=None)):
+    """List repositories in the configured Gitea organization and whether they are already onboarded."""
+    gitea_api = _create_gitea_api()
+    resolved_owner = owner or database_module.settings.gitea_owner
+    repos = gitea_api.list_repositories_sync(owner=resolved_owner)
+    if isinstance(repos, dict) and repos.get('error'):
+        raise HTTPException(status_code=502, detail=repos.get('error'))
+    manager = DatabaseManager(db)
+    items = []
+    for repo in repos if isinstance(repos, list) else []:
+        tracked_project = manager.get_project_by_repository(resolved_owner, repo.get('name', ''))
+        items.append(
+            {
+                'name': repo.get('name'),
+                'full_name': repo.get('full_name') or f"{resolved_owner}/{repo.get('name')}",
+                'description': repo.get('description'),
+                'html_url': repo.get('html_url'),
+                'clone_url': repo.get('clone_url'),
+                'default_branch': repo.get('default_branch'),
+                'private': bool(repo.get('private', False)),
+                'onboarded': tracked_project is not None,
+                'project_id': tracked_project.project_id if tracked_project is not None else None,
+            }
+        )
+    return {'repositories': items}
+
+
+@app.post('/gitea/repos/onboard')
+async def onboard_gitea_repository(request: GiteaRepositoryOnboardRequest, db: DbSession):
+    """Onboard a manually created Gitea repository into the factory dashboard."""
+    gitea_api = _create_gitea_api()
+    owner = request.owner or database_module.settings.gitea_owner
+    repo = await gitea_api.get_repo_info(owner=owner, repo=request.repo_name)
+    if isinstance(repo, dict) and repo.get('error'):
+        raise HTTPException(status_code=404, detail=repo.get('error'))
+    manager = DatabaseManager(db)
+    onboarded = manager.onboard_repository(owner=owner, repo_name=request.repo_name, repository_data=repo)
+    manager.sync_repository_issues(project_id=onboarded['project_id'], gitea_api=gitea_api, state='open')
+    sync_result = None
+    if request.sync_commits:
+        sync_result = manager.sync_repository_activity(
+            project_id=onboarded['project_id'],
+            gitea_api=gitea_api,
+            commit_limit=request.commit_limit,
+        )
+    return {
+        'status': 'success',
+        'onboarded': onboarded,
+        'sync_result': sync_result,
+    }
+
+
 @app.get('/audit/logs')
 def get_audit_logs(db: DbSession):
     """Return all project logs ordered newest first."""
@@ -348,7 +873,7 @@ async def setup_n8n_workflow(request: N8NSetupRequest, db: DbSession):
     )
     result = await agent.setup(
         webhook_path=request.webhook_path,
-        backend_url=request.backend_url or f"{database_module.settings.backend_public_url}/generate",
+        backend_url=request.backend_url or f"{database_module.settings.backend_public_url}/generate/text",
         force_update=request.force_update,
         telegram_bot_token=database_module.settings.telegram_bot_token,
         telegram_credential_name=database_module.settings.n8n_telegram_credential_name,