Commit Graph

218 Commits

Author SHA1 Message Date
5639d085b4 feat(#154): Implement context estimator
Implements formula-based context estimation for predicting token
usage before issue assignment.

Formula:
  base = (files × 7000) + complexity + tests + docs
  total = base × 1.3  (30% safety buffer)

Features:
- EstimationInput/Result data models with validation
- ComplexityLevel, TestLevel, DocLevel enums
- Agent recommendation (haiku/sonnet/opus) based on tokens
- Validation against actual usage with tolerance checking
- Convenience function for quick estimations
- JSON serialization support

Implementation:
- issue_estimator.py: Core estimator with formula
- models.py: Data models and enums (100% coverage)
- test_issue_estimator.py: 35 tests, 100% coverage
- ESTIMATOR.md: Complete API documentation
- requirements.txt: Python dependencies
- .coveragerc: Coverage configuration

Test Results:
- 35 tests passing
- 100% code coverage (excluding __main__)
- Validates against historical issues
- All edge cases covered

Acceptance Criteria Met:
 Context estimation formula implemented
 Validation suite tests against historical issues
 Formula includes all components (files, complexity, tests, docs, buffer)
 Unit tests for estimator (100% coverage, exceeds 85% requirement)
 All components tested (low/medium/high levels)
 Agent recommendation logic validated

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-02-01 17:42:59 -06:00
e23c09f1f2 feat(#157): Set up webhook receiver endpoint
Implement FastAPI webhook receiver for Gitea issue assignment events
with HMAC SHA256 signature verification and event routing.

Implementation details:
- FastAPI application with /webhook/gitea POST endpoint
- HMAC SHA256 signature verification in security.py
- Event routing for assigned, unassigned, closed actions
- Comprehensive logging for all webhook events
- Health check endpoint at /health
- Docker containerization with health checks
- 91% test coverage (exceeds 85% requirement)

TDD workflow followed:
- Wrote 16 tests first (RED phase)
- Implemented features to pass tests (GREEN phase)
- All tests passing with 91% coverage
- Type checking with mypy: success
- Linting with ruff: success

Files created:
- apps/coordinator/src/main.py - FastAPI application
- apps/coordinator/src/webhook.py - Webhook handlers
- apps/coordinator/src/security.py - HMAC verification
- apps/coordinator/src/config.py - Configuration management
- apps/coordinator/tests/ - Comprehensive test suite
- apps/coordinator/Dockerfile - Production container
- apps/coordinator/pyproject.toml - Python project config

Configuration:
- Updated .env.example with GITEA_WEBHOOK_SECRET
- Updated docker-compose.yml with coordinator service

Testing:
- 16 unit and integration tests
- Security tests for signature verification
- Event handler tests for all supported actions
- Health check endpoint tests
- All tests passing with 91% coverage

This unblocks issue #158 (issue parser).

Fixes #157

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-02-01 17:41:46 -06:00
658ec0774d fix(ci): Switch to Kaniko for daemonless container builds
All checks were successful
ci/woodpecker/push/woodpecker Pipeline was successful
docker:dind requires privileged mode and a running daemon.
Kaniko builds containers without needing Docker daemon:
- Runs unprivileged
- Reads credentials from /kaniko/.docker/config.json
- Designed for CI environments like Woodpecker

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-02-01 17:34:50 -06:00
de3f3b9204 feat(#156): Create coordinator bot user documentation and setup scripts
Add comprehensive documentation and automated scripts for setting up the mosaic
coordinator bot user in Gitea. This enables the coordinator system to manage
issue assignments, comments, and orchestration.

Changes:
- docs/1-getting-started/3-configuration/4-gitea-coordinator.md: Complete setup guide
  * Step-by-step bot user creation via UI and API
  * Repository permission configuration
  * API token generation and storage
  * Comprehensive testing procedures
  * Security best practices and troubleshooting

- scripts/coordinator/create-gitea-bot.sh: Automated bot creation script
  * Creates mosaic bot user with proper configuration
  * Sets up repository permissions
  * Generates API token
  * Tests authentication
  * Provides credential output for secure storage

- scripts/coordinator/test-gitea-bot.sh: Bot functionality test suite
  * Tests authentication
  * Verifies repository access
  * Tests issue operations (read, list, assign, comment)
  * Validates label management
  * Confirms all required permissions

- scripts/coordinator/README.md: Scripts usage documentation
  * Workflow guides
  * Configuration reference
  * Troubleshooting section
  * Token rotation procedures

- .env.example: Added Gitea coordinator configuration template
  * GITEA_URL, GITEA_BOT_USERNAME, GITEA_BOT_TOKEN
  * GITEA_BOT_PASSWORD, GITEA_REPO_OWNER, GITEA_REPO_NAME
  * Security notes for credential storage

All acceptance criteria met:
✓ Documentation for bot user creation
✓ Automated setup script
✓ Testing procedures and scripts
✓ Configuration templates
✓ Security best practices
✓ Troubleshooting guide

Addresses Milestone: M4.1-Coordinator
Relates to: #140, #157, #158

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-02-01 17:32:03 -06:00
32c35d327b fix(ci): Use docker:dind with manual login instead of buildx plugin
Some checks failed
ci/woodpecker/push/woodpecker Pipeline failed
The buildx plugin's credential handling doesn't work properly with
Harbor. The docker-auth-test step proved that standard docker login
works, so we switch to:
- docker:dind image
- Manual docker login before build
- Standard docker build and docker push

This bypasses buildx's separate credential store issue.

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-02-01 17:31:05 -06:00
211c532fb0 fix(ci): Add auth debug step, switch back to buildx
Some checks failed
ci/woodpecker/push/woodpecker Pipeline failed
Added a docker-auth-test step that:
- Shows credential lengths (for debugging)
- Tests docker login directly with Harbor

This will help identify if the issue is with secrets injection
or with how buildx handles authentication.

Reverted to woodpeckerci/plugin-docker-buildx since plugins/docker
requires server-side WOODPECKER_PLUGINS_PRIVILEGED config.

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-02-01 17:24:51 -06:00
b1be63edd6 fix(ci): Correct repo path format for plugins/docker
The repo setting should NOT include the registry prefix - the
registry setting handles that separately.

Changed repo: reg.mosaicstack.dev/mosaic/api -> repo: mosaic/api

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-02-01 17:23:49 -06:00
da038d3df2 fix(ci): Switch from buildx to plugins/docker for Harbor auth
The woodpeckerci/plugin-docker-buildx plugin was failing with
"insufficient_scope: authorization failed" when pushing to Harbor,
even though the same credentials worked locally.

Switched to the standard plugins/docker which uses traditional
docker login authentication that may work better with Harbor.

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-02-01 17:13:58 -06:00
e1ed98b038 fix: Remove privileged flag (not allowed), keep debug
Some checks failed
ci/woodpecker/push/woodpecker Pipeline failed
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-02-01 16:18:42 -06:00
55b2ddb58a fix: Add privileged and debug flags to docker-buildx steps
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-02-01 16:18:15 -06:00
8ca0b45fcb fix: Allow docker builds on manual pipeline triggers
Some checks failed
ci/woodpecker/push/woodpecker Pipeline failed
ci/woodpecker/manual/woodpecker Pipeline failed
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-02-01 15:10:59 -06:00
cd727f619f feat: Add debug output to Dockerfiles and .dockerignore
Some checks failed
ci/woodpecker/push/woodpecker Pipeline failed
ci/woodpecker/manual/woodpecker Pipeline was successful
- Add .dockerignore to exclude node_modules, dist, and build artifacts
- Add pre/post build directory listings to diagnose dist not found issue
- Disable turbo cache temporarily with --force flag
- Add --verbosity=2 for more detailed turbo output

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-02-01 14:50:13 -06:00
763409cbb4 fix: Remove registry prefix from repo paths in Woodpecker
Some checks failed
ci/woodpecker/push/woodpecker Pipeline failed
The docker-buildx plugin automatically prepends registry to repo,
so having the full URL caused doubled paths:
reg.mosaicstack.dev/reg.mosaicstack.dev/mosaic/api

Changed from: repo: reg.mosaicstack.dev/mosaic/api
Changed to:   repo: mosaic/api

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-02-01 13:45:29 -06:00
45483934c3 Merge branch 'fix/harbor-registry-url' into develop
Some checks failed
ci/woodpecker/push/woodpecker Pipeline failed
2026-02-01 13:39:38 -06:00
442c2f7de2 fix: Dockerfile COPY order - node_modules must come after source
All checks were successful
ci/woodpecker/push/woodpecker Pipeline was successful
Docker COPY replaces directory contents, so copying source code
after node_modules was wiping the deps. Reordered to:
1. Copy source code first
2. Copy node_modules second (won't be overwritten)

Fixes API build failure: "dist not found"

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-02-01 13:39:25 -06:00
728f68f877 Merge pull request 'fix(ci): Update Harbor registry URL to reg.mosaicstack.dev' (#178) from fix/harbor-registry-url into develop
Some checks failed
ci/woodpecker/push/woodpecker Pipeline failed
2026-02-01 19:26:17 +00:00
365975d76e fix(ci): Update Harbor registry URL to reg.mosaicstack.dev
Some checks failed
ci/woodpecker/push/woodpecker Pipeline failed
ci/woodpecker/pr/woodpecker Pipeline failed
Changed from reg.diversecanvas.com to reg.mosaicstack.dev

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-02-01 13:25:55 -06:00
1bfdd57f04 Merge pull request 'Release: CI/CD Pipeline & Architecture Updates' (#177) from develop into main
Some checks failed
ci/woodpecker/push/woodpecker Pipeline failed
Reviewed-on: #177
2026-02-01 19:18:47 +00:00
4b943fb997 feat: Add Docker build & push to Woodpecker CI pipeline
All checks were successful
ci/woodpecker/manual/woodpecker Pipeline was successful
ci/woodpecker/pr/woodpecker Pipeline was successful
- Add docker-build-api, docker-build-web, docker-build-postgres steps
- Images pushed to reg.diversecanvas.com/mosaic/* on main/develop
- Create docker-compose.prod.yml for production deployments
- Add .env.prod.example with production configuration

Requires Harbor secrets in Woodpecker:
- harbor_username
- harbor_password

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-02-01 01:50:02 -06:00
9246f56687 fix(api): Add AuthModule import to modules using AuthGuard
All checks were successful
ci/woodpecker/push/woodpecker Pipeline was successful
Modules using AuthGuard in their controllers need to import AuthModule
to make AuthService available for dependency injection.

Fixed:
- ActivityModule
- WorkspaceSettingsModule

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-02-01 01:48:09 -06:00
fb0f6b5b62 fix(docker): Fix module resolution and healthcheck syntax
All checks were successful
ci/woodpecker/push/woodpecker Pipeline was successful
Issues fixed:
1. Module not found: Added missing copy of apps/{api,web}/node_modules
   which contains pnpm symlinks to the root node_modules

2. Healthcheck syntax: Fixed broken quoting from prettier reformatting
   Changed to CMD-SHELL with proper escaping

3. Removed obsolete version: "3.9" from docker-compose.yml

The apps need their own node_modules directories because pnpm uses
symlinks that point from apps/*/node_modules to node_modules/.pnpm/*

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-02-01 01:37:30 -06:00
aa17b9cb3b fix(docker): Make port configuration consistent and dynamic
All checks were successful
ci/woodpecker/push/woodpecker Pipeline was successful
Fixed the mismatch between environment variables:
- docker-compose now passes PORT (what NestJS/Next.js read) instead of API_PORT
- API_PORT/WEB_PORT control host mapping, PORT controls container

Changes:
- docker-compose: Pass PORT=${API_PORT} and PORT=${WEB_PORT} to containers
- docker-compose: Dynamic port mapping on both host and container sides
- docker-compose: Traefik labels use ${API_PORT}/${WEB_PORT} variables
- docker-compose: Healthchecks use PORT env var
- Dockerfiles: Removed hardcoded port values
- Dockerfiles: Healthchecks read PORT at runtime

This allows changing ports via API_PORT/WEB_PORT environment variables
and have all components (app, healthcheck, Traefik) use the correct port.

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-02-01 01:29:15 -06:00
8f63b3e1dc docs: Add Mosaic Component Architecture and Guard Rails design docs
All checks were successful
ci/woodpecker/push/woodpecker Pipeline was successful
- mosaic-component-architecture.md: OpenClaw wrapper pattern, component naming,
  job tracking, chat integration, database schema
- guard-rails-capability-permissions.md: Capability-based permission model

Related: #162 (M4.2 Infrastructure Epic)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-02-01 01:26:01 -06:00
e045cb5a45 perf(docker): Add BuildKit cache mounts for faster builds
All checks were successful
ci/woodpecker/push/woodpecker Pipeline was successful
Added cache mounts for:
- pnpm store: Caches downloaded packages between builds
- TurboRepo: Caches build outputs between builds

This significantly speeds up subsequent builds:
- First build: Full download and compile
- Subsequent builds: Only changed packages are re-downloaded/rebuilt

Requires Docker BuildKit (default in Docker 23+).

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-02-01 01:22:51 -06:00
353f04f950 fix(docker): Ensure public directory exists in web builder
All checks were successful
ci/woodpecker/push/woodpecker Pipeline was successful
The production stage was failing because it tried to copy the public
directory which doesn't exist in the source. Added mkdir -p to ensure
the directory exists (even if empty) before the production stage
tries to copy it.

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-02-01 01:15:34 -06:00
38f22f0b4e fix(scripts): Improve base URL configuration display clarity
All checks were successful
ci/woodpecker/push/woodpecker Pipeline was successful
When detecting existing configuration, the setup script now shows a
detailed breakdown instead of just "Current base URL: ...":

  Mode:    Traefik reverse proxy

  Web URL: https://app.mosaicstack.dev
  API URL: https://api.mosaicstack.dev
  Auth:    https://auth.mosaicstack.dev

This makes it clear:
- What access mode is configured (localhost/IP/domain/Traefik)
- What each URL is used for (Web UI, API, Authentication)
- Whether to change the configuration

Added helper functions:
- detect_access_mode(): Determines mode from existing .env values
- display_access_config(): Formats the URL breakdown display

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-02-01 00:57:23 -06:00
0495c48418 fix(docker): Copy node_modules from builder instead of reinstalling
All checks were successful
ci/woodpecker/push/woodpecker Pipeline was successful
pnpm stores the Prisma client in the content-addressable store at
node_modules/.pnpm/.../.prisma, not at apps/api/node_modules/.prisma.
The production stage was trying to copy from the wrong location.

Additionally, running `pnpm install --prod` in production failed because:
1. The husky prepare script runs but husky is a devDependency
2. The Prisma client postinstall can't run without the prisma CLI

Fixed by copying the full node_modules from the builder stage, which
already has all dependencies properly installed and the Prisma client
generated in the correct pnpm store location.

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-02-01 00:42:34 -06:00
7ee08865fd fix(docker): Use TurboRepo to build workspace dependencies
All checks were successful
ci/woodpecker/push/woodpecker Pipeline was successful
The Docker builds were failing because they ran `pnpm build` directly
in the app directories without first building workspace dependencies
(@mosaic/shared, @mosaic/ui). CI passed because it runs TurboRepo
from the root which respects the dependency graph.

Changed both Dockerfiles to use `pnpm turbo build --filter=@mosaic/{app}`
which ensures dependencies are built in the correct order:
- Web: @mosaic/config → @mosaic/shared → @mosaic/ui → @mosaic/web
- API: @mosaic/config → @mosaic/shared → prisma:generate → @mosaic/api

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-02-01 00:37:34 -06:00
a84d06815e fix(docker): Make prepare script work in production builds
All checks were successful
ci/woodpecker/push/woodpecker Pipeline was successful
The husky prepare script was failing during Docker production builds
because husky is a devDependency and isn't available when running
`pnpm install --prod --frozen-lockfile`.

Changed from `husky install` (deprecated in v9+) to `husky || true`
which gracefully handles the case when husky isn't installed.

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-02-01 00:30:37 -06:00
8c8d065cc2 feat(arch): Add Guard Rails capability-based permission system design
All checks were successful
ci/woodpecker/push/woodpecker Pipeline was successful
Guard Rails complement Quality Rails by controlling what agents can do:
- Capability-based permissions (resource:action pattern)
- Read/organize/draft allowed by default
- Execute/admin require explicit grants
- Human-in-the-loop approval for sensitive actions

Examples: email (read/draft , send ), git (commit , force push )

Also:
- Add .admin-credentials and .env.bak.* to .gitignore

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-02-01 00:25:53 -06:00
98f80eaf51 fix(scripts): Fix awk env parsing for POSIX compatibility
Some checks failed
ci/woodpecker/push/woodpecker Pipeline failed
- Use index() instead of regex capture groups for key extraction
- More portable across different awk implementations

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-02-01 00:24:31 -06:00
e63c19d158 chore: Cleanup QA reports and improve setup scripts
All checks were successful
ci/woodpecker/push/woodpecker Pipeline was successful
Scripts:
- common.sh: Fix select_option to use /dev/tty for interactive prompts
- common.sh: Improve check_docker with detailed error messages
- setup.sh: Add Traefik configuration options
- setup.sh: Add argument validation for --mode, --external-authentik, etc.
- setup.sh: Add fun taglines

QA Reports:
- Remove stale remediation reports
- Keep current pending reports

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-01-31 22:53:47 -06:00
cb0948214e feat(auth): Configure Authentik OIDC integration with better-auth
All checks were successful
ci/woodpecker/push/woodpecker Pipeline was successful
- Add genericOAuth plugin to auth.config.ts with Authentik provider
- Fix LoginButton to use /auth/signin/authentik (not /auth/callback/)
- Add production URLs to trustedOrigins
- Update .env.example with correct redirect URI documentation

Redirect URI for Authentik: https://api.mosaicstack.dev/auth/callback/authentik

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-01-31 18:11:32 -06:00
f2b25079d9 fix(#27): address security issues in intent classification
All checks were successful
ci/woodpecker/push/woodpecker Pipeline was successful
- Add input sanitization to prevent LLM prompt injection
  (escapes quotes, backslashes, replaces newlines)
- Add MaxLength(500) validation to DTO to prevent DoS
- Add entity validation to filter malicious LLM responses
- Add confidence validation to clamp values to 0.0-1.0
- Make LLM model configurable via INTENT_CLASSIFICATION_MODEL env var
- Add 12 new security tests (total: 72 tests, from 60)

Security fixes identified by code review:
- CVE-mitigated: Prompt injection via unescaped user input
- CVE-mitigated: Unvalidated entity data from LLM response
- CVE-mitigated: Missing input length validation

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-01-31 16:50:32 -06:00
fd93be6032 feat: Add comprehensive setup wizard foundation
Modeled after Calibr setup.sh pattern (~/src/calibr/scripts/setup.sh).

Implemented (Foundation):
- Platform detection (Ubuntu, Arch, macOS, Fedora)
- Dependency checking and installation
- Mode selection (Docker vs Native)
- Interactive + non-interactive modes
- Comprehensive logging (clean console + full trace to log file)
- Common utility functions library (450+ lines)

Features in common.sh:
- Output formatting (colors, headers, success/error/warning)
- User input (confirm, select_option)
- Platform detection
- Dependency checking (Docker, Node, pnpm, PostgreSQL)
- Package installation (apt, pacman, dnf, brew)
- Validation (URL, email, port, domain)
- Secret generation (cryptographically secure)
- .env file parsing and management
- Port conflict detection
- File backup with timestamps

To Be Implemented (See scripts/README.md):
- Complete configuration collection
- .env generation with smart preservation
- Port conflict detection
- Password/secret generation
- Authentik blueprint auto-configuration
- Docker deployment execution
- Post-install instructions

Usage:
  ./scripts/setup.sh                    # Interactive
  ./scripts/setup.sh --help             # Show options
  ./scripts/setup.sh --dry-run          # Preview
  ./scripts/setup.sh --non-interactive  # CI/CD

Refs: Setup wizard issue (created)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-01-31 16:45:56 -06:00
0eb3abc12c Clean up documents located in the project root.
All checks were successful
ci/woodpecker/push/woodpecker Pipeline was successful
2026-01-31 16:42:26 -06:00
d7f04d1148 feat(#27): implement intent classification service
All checks were successful
ci/woodpecker/push/woodpecker Pipeline was successful
Implement intent classification for natural language queries in the brain module.

Features:
- Hybrid classification approach: rule-based (fast, <100ms) with optional LLM fallback
- 10 intent types: query_tasks, query_events, query_projects, create_task, create_event, update_task, update_event, briefing, search, unknown
- Entity extraction: dates, times, priorities, statuses, people
- Pattern-based matching with priority system (higher priority = checked first)
- Optional LLM classification for ambiguous queries
- POST /api/brain/classify endpoint

Implementation:
- IntentClassificationService with classify(), classifyWithRules(), classifyWithLlm(), extractEntities()
- Comprehensive regex patterns for common query types
- Entity extraction for dates, times, priorities, statuses, mentions
- Type-safe interfaces for IntentType, IntentClassification, ExtractedEntity, IntentPattern
- ClassifyIntentDto and IntentClassificationResultDto for API validation
- Integrated with existing LlmService (optional dependency)

Testing:
- 60 comprehensive tests covering all intent types
- Edge cases: empty queries, special characters, case sensitivity, multiple whitespace
- Entity extraction tests with position tracking
- LLM fallback tests with error handling
- 100% test coverage
- All tests passing (60/60)
- TDD approach: tests written first

Quality:
- No explicit any types
- Explicit return types on all functions
- No TypeScript errors
- Build successful
- Follows existing code patterns
- Quality Rails compliance: All lint checks pass

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-01-31 15:41:10 -06:00
403aba4cd3 docs: Add issue parser estimation strategy
Critical enhancement for real-world usage - parser must handle:
- Unformatted issues (estimate from content)
- Incomplete metadata (best-guess + confidence score)
- Oversized issues (auto-decompose before queuing)

Three-level estimation:
1. Structured metadata → extract directly (95%+ confidence)
2. Content analysis → AI estimates from description (50-95%)
3. Minimal info → defaults + warn user (<50%)

50% rule enforcement:
- Detect issues > 50% of agent's context limit
- Auto-decompose into sub-issues using Opus
- Create sub-issues in Gitea with dependencies
- Label parent as EPIC

Confidence-based workflow:
- ≥60%: Queue automatically
- 30-59%: Queue with warning
- <30%: Don't queue, request more details

Makes coordinator truly autonomous - handles whatever users throw at it.

Refs #158 (COORD-002)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-01-31 15:40:34 -06:00
3be60ccd18 docs: Add assignment-based trigger architecture
Implements Phase 0 foundation for non-AI coordinator.

Key features:
- User assigns issue to @mosaic bot user → triggers webhook
- Webhook receiver processes assignment events
- AI agent parses issue metadata (context, difficulty, agent)
- Queue manager tracks dependencies and status
- Orchestration loop spawns agents and monitors progress

Benefits:
- Natural Gitea workflow (just assign issues)
- Visual feedback in Gitea UI
- Granular control (assign what you want)
- Event-driven (webhooks, not polling)
- No CLI needed

Phase 0 issues: #156-161 (6 issues, 290.6K tokens)

Refs #142

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-01-31 15:30:21 -06:00
3d6159ae15 fix: address code review issues and cleanup QA reports
All checks were successful
ci/woodpecker/push/woodpecker Pipeline was successful
Code review fixes:
- Add error logging to LlmProviderAdminController.testProvider catch block
- Use atomic increment operations in TokenBudgetService.updateUsage to prevent race conditions
- Update test expectations for atomic increment pattern

Cleanup:
- Remove obsolete QA automation reports

All 1169 tests passing.

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-01-31 15:01:18 -06:00
903109ea40 docs: Add overlap analysis for non-AI coordinator patterns
All checks were successful
ci/woodpecker/push/woodpecker Pipeline was successful
Detailed comparison showing:
- Existing doc addresses L-015 (premature completion)
- New doc addresses context exhaustion (multi-issue orchestration)
- ~20% overlap (both use non-AI coordinator, mechanical gates)
- 80% complementary (different problems, different solutions)

Recommends merging into comprehensive document (already done).

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-01-31 14:47:59 -06:00
a2f06fe75b docs: Add comprehensive non-AI coordinator architecture
Merges two complementary architectural patterns:
1. Quality Enforcement Layer - Prevents premature agent completion
2. Orchestration Layer - Manages multi-agent context and assignment

Key features:
- 50% rule for issue sizing
- Agent profiles and cost optimization
- Context monitoring (compact at 80%, rotate at 95%)
- Mechanical quality gates (build, lint, test, coverage)
- Forced continuation when gates fail
- 4-week PoC plan

Addresses issue #140 and L-015 (Agent Premature Completion)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-01-31 14:47:09 -06:00
4b4d21c732 feat(#129): add LLM provider admin API endpoints
Implement REST API endpoints for managing LLM provider instances.

Changes:
- Created DTOs for provider CRUD operations (CreateLlmProviderDto, UpdateLlmProviderDto, LlmProviderResponseDto)
- Implemented LlmProviderAdminController with full CRUD endpoints:
  - GET /llm/admin/providers - List all providers
  - GET /llm/admin/providers/:id - Get provider details
  - POST /llm/admin/providers - Create new provider
  - PATCH /llm/admin/providers/:id - Update provider
  - DELETE /llm/admin/providers/:id - Delete provider
  - POST /llm/admin/providers/:id/test - Test connection
  - POST /llm/admin/reload - Reload from database
- Updated llm-manager.service.ts to support OpenAI and Claude providers
- Added comprehensive test suite with 97.95% coverage
- Proper validation, error handling, and type safety

All tests pass. Pre-commit hooks pass.

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-01-31 14:37:55 -06:00
772776bfd9 feat(#125): add Claude (Anthropic) LLM provider
Implement Anthropic Claude provider for Claude Opus, Sonnet, and Haiku models.

Implementation details:
- Created ClaudeProvider class implementing LlmProviderInterface
- Added @anthropic-ai/sdk npm package integration
- Implemented chat completion with streaming support
- Claude-specific message format (system prompt separate from messages)
- Static model list (Claude API doesn't provide list models endpoint)
- Embeddings throw error as Claude doesn't support native embeddings
- Added OpenTelemetry tracing with @TraceLlmCall decorator
- 100% statement, function, and line coverage (79% branch coverage)

Tests:
- Created comprehensive test suite with 20 tests
- All tests follow TDD pattern (written before implementation)
- Tests cover initialization, health checks, chat, streaming, and error handling
- Mocked Anthropic SDK client for isolated unit testing

Quality checks:
- All tests pass (1131 total tests across project)
- ESLint passes with no errors
- TypeScript type checking passes
- Follows existing code patterns from OpenAI and Ollama providers

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-01-31 14:29:40 -06:00
0fdcfa6ed3 feat(#124): add OpenAI LLM provider
Implement OpenAI provider for GPT-4, GPT-3.5, and other OpenAI models.

Implementation includes:
- OpenAI SDK integration with API key authentication
- Chat completion with streaming support
- Embeddings generation
- Health checks and model listing
- OpenTelemetry tracing
- Comprehensive test suite with 97% coverage

Follows TDD methodology:
- Written tests first (RED phase)
- Implemented minimal code to pass tests (GREEN phase)
- Code passes typecheck, linter, and all quality gates

Test coverage: 97.18% statements, 97.05% lines
All 22 tests passing

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-01-31 14:21:38 -06:00
faf6328e0b test(#141): add Non-AI Coordinator integration tests
Comprehensive E2E validation proving coordinator enforces quality
gates and prevents premature completion claims.

Test scenarios (21 tests):
- Rejection Flow: Build/lint/test/coverage gate failures
- Acceptance Flow: All gates pass, required-only pass
- Continuation Flow: Retry, escalation, attempt tracking
- Escalation Flow: Manual review, notifications, history
- Configuration: Workspace-specific, defaults, custom gates
- Performance: Timeout compliance, memory limits
- Complete E2E: Full rejection-continuation-acceptance cycle

Fixtures:
- mock-agent-outputs.ts: Simulated gate execution results
- mock-gate-configs.ts: Various gate configurations

Validates integration of:
- Quality Orchestrator (#134)
- Quality Gate Config (#135)
- Completion Verification (#136)
- Continuation Prompts (#137)
- Rejection Handler (#139)

All 21 tests passing

Fixes #141

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-01-31 14:14:56 -06:00
a86d304f07 feat(#139): build Gate Rejection Response Handler
All checks were successful
ci/woodpecker/push/woodpecker Pipeline was successful
Implement rejection handling for tasks that fail quality gates after
all continuation attempts are exhausted.

Schema:
- Add TaskRejection model for tracking rejections
- Store failures, attempts, escalation state

Service:
- handleRejection: Main entry point for rejection handling
- logRejection: Database logging
- determineEscalation: Rule-based escalation determination
- executeEscalation: Execute escalation actions
- sendNotification: Notification dispatch
- markForManualReview: Flag tasks for human review
- getRejectionHistory: Query rejection history
- generateRejectionReport: Markdown report generation

Escalation rules:
- max-attempts: Trigger after 3+ attempts
- time-exceeded: Trigger after 2+ hours
- critical-failure: Trigger on security/critical issues

Actions: notify, block, reassign, cancel

Tests: 16 passing with 80% statement coverage

Fixes #139

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-01-31 14:01:42 -06:00
0387cce116 feat(#137): create Forced Continuation Prompt System
Implement prompt generation system that produces continuation prompts
based on verification failures to force AI agents to complete work.

Service:
- generatePrompt: Complete prompt from failure context
- generateTestFailurePrompt: Test-specific guidance
- generateBuildErrorPrompt: Build error resolution
- generateCoveragePrompt: Coverage improvement strategy
- generateIncompleteWorkPrompt: Completion requirements

Templates:
- base.template: System/user prompt structure
- test-failure.template: Test fix guidance
- build-error.template: Compilation error guidance
- coverage.template: Coverage improvement strategy
- incomplete-work.template: Completion requirements

Constraint escalation:
- Attempt 1: Normal guidance
- Attempt 2: Focus only on failures
- Attempt 3: Minimal changes only
- Final: Last attempt warning

Priority levels: critical/high/normal based on failure severity

Tests: 24 passing with 95.31% coverage

Fixes #137

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-01-31 13:51:46 -06:00
72ae92f5a6 feat(#136): build Completion Verification Engine
Implement verification engine to determine if AI agent work is truly
complete by analyzing outputs and detecting deferred work patterns.

Strategies:
- FileChangeStrategy: Detect TODO/FIXME, placeholders, stubs
- TestOutputStrategy: Validate pass rates, coverage (85%), skipped tests
- BuildOutputStrategy: Detect TS errors, ESLint errors, build failures

Deferred work detection patterns:
- "follow-up", "to be added later"
- "incremental improvement", "future enhancement"
- "TODO: complete", "placeholder implementation"
- "stub", "work in progress", "partially implemented"

Features:
- Confidence scoring (0-100%)
- Verdict system: complete/incomplete/needs-review
- Actionable suggestions for improvements
- Strategy-based extensibility

Integration:
- Complements Quality Orchestrator (#134)
- Uses Quality Gate Config (#135)

Tests: 46 passing with 95.27% coverage

Fixes #136

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-01-31 13:44:23 -06:00
4a2909ce1e feat(#135): implement Quality Gate Configuration System
Add database-backed quality gate configuration for workspaces with
full CRUD operations and default gate seeding.

Schema:
- Add QualityGate model with workspace relation
- Support for custom commands and regex patterns
- Enable/disable and ordering support

Service:
- CRUD operations for quality gates
- findEnabled: Get ordered, enabled gates
- reorder: Bulk reorder with transaction
- seedDefaults: Seed 4 default gates
- toOrchestratorFormat: Convert to orchestrator interface

Endpoints:
- GET /workspaces/:id/quality-gates - List
- GET /workspaces/:id/quality-gates/:gateId - Get one
- POST /workspaces/:id/quality-gates - Create
- PATCH /workspaces/:id/quality-gates/:gateId - Update
- DELETE /workspaces/:id/quality-gates/:gateId - Delete
- POST /workspaces/:id/quality-gates/reorder
- POST /workspaces/:id/quality-gates/seed-defaults

Default gates: Build, Lint, Test, Coverage (85%)

Tests: 25 passing with 95.16% coverage

Fixes #135

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-01-31 13:33:04 -06:00