centient 2.20.0

package/CHANGELOG.md

@@ -0,0 +1,362 @@
+# Changelog
+
+All notable changes to the MCP Server for Context Engineering.
+
+The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
+and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
+
+## [2.20.0] - 2026-01-11
+
+### Added
+
+- **npm Publishing Support** - Package ready for npm registry
+  - Added `files` field to package.json (dist, README.md, CHANGELOG.md)
+  - Added repository, bugs, homepage metadata
+  - npx installation now supported: `npx centient`
+
+### Changed
+
+- **README.md** - Simplified installation with three options:
+  - Option 1: npx (recommended, zero-install)
+  - Option 2: Global install via npm
+  - Option 3: From source (for developers)
+- Updated tool count to 84 across 15 modules
+- Standardized repository URLs to Szermer/centient
+
+### Documentation
+
+- Installation friction analysis completed
+- API key landscape documented (4 services, 5 env vars)
+- Feature tiers clarified (core works without API keys)
+
+## [2.19.0] - 2026-01-09
+
+### Added
+
+- **OpenRouter Integration** - Unified LLM API for consolidated billing
+  - Route all chat completions through OpenRouter (300+ models available)
+  - Support for Anthropic models: claude-sonnet-4, claude-opus-4, claude-haiku-3.5
+  - Support for DeepSeek models: deepseek-chat, deepseek-reasoner
+  - Support for Meta models: llama-3.1-70b-instruct, llama-3.1-405b-instruct
+  - Model ID mapping for backwards compatibility (e.g., `gpt-4o` → `openai/gpt-4o`)
+  - OpenRouter pricing added to `costPricing.ts`
+  - Circuit breaker and rate limiter for OpenRouter service
+  - Health monitoring tools updated to include OpenRouter
+
+### Changed
+
+- **llmStreamClient.ts** - Rewrote to use OpenRouter API (OpenAI SDK compatible)
+- **Environment variables** - Clarified purpose of each API key in `.env.example`:
+  - `OPENROUTER_API_KEY` - LLM completions (unified billing)
+  - `OPENAI_API_KEY` - Embeddings only (Qdrant vector storage)
+  - `GEMINI_API_KEY` - File Search only (semantic artifact search)
+- Health tools now report 4 circuit breaker services and 6 rate limiter services
+
+## [2.14.0] - 2025-01-02
+
+### Added
+
+- **`get_session_summary` tool** - Comprehensive session summary with branch visualization
+  - ASCII tree visualization showing decision points and branches
+  - Status icons: ★ active, ✓ merged, ✗ rejected, ○ abandoned
+  - Formatted memories as markdown (decisions, hypotheses, learnings)
+  - Configurable via `includeTree` and `includeMemories` options
+- 22 new tests for session summary functionality
+
+### Changed
+
+- Session module now has 19 tools (was 18)
+- Total tools: 71 across 14 modules
+- Test suite: 801 tests (was 779)
+
+## [2.13.0] - 2025-01-02
+
+### Added
+
+- **Session Branching Module** (6 new tools)
+  - `mark_decision_point` - Mark where multiple alternatives exist
+  - `list_decision_points` - List all decision points with branch counts
+  - `create_branch` - Create exploration branch from decision point
+  - `list_branches` - View branches with ASCII tree visualization
+  - `switch_branch` - Switch active branch context
+  - `close_branch` - Close via merge/reject/abandon
+- Branch lifecycle: active → merged/rejected/abandoned
+- Decision point categories: architecture, implementation, tooling, refactoring, exploration, integration
+- 38 new tests for branching functionality
+
+### Changed
+
+- Total tools: 70 across 14 modules (was 64 across 13 modules)
+- Test suite: 779 tests (was 741)
+
+## [2.12.0] - 2025-01-02
+
+### Added
+
+- **Hallucination Detection Layer** (6 new tools in Session module)
+  - `check_claim_consistency` - Cross-reference claims against knowledge stores
+  - `validate_citation` - Verify cited references exist and support claims
+  - `get_verification_prompt` - Generate tailored verification questions
+  - `flag_for_verification` - Flag low-confidence content for human review
+  - `record_verification_outcome` - Record human verification results
+  - `check_code_feasibility` - Validate code syntax, imports, and security
+- Confidence badges: ✅ VERIFIED, ⚠️ UNVERIFIED, 🧪 EXPERIMENTAL
+- Learning loop for hallucination detection accuracy
+
+### Changed
+
+- Session module expanded from 12 to 18 tools
+- Added ADR-017: Hallucination Detection Layer
+
+## [2.11.0] - 2025-12-30
+
+### Added
+
+- **Pattern Versioning** (3 new tools)
+  - `getPatternVersions` - List versions with changelogs
+  - `createPatternVersion` - Create new semver version
+  - `deprecatePatternVersion` - Mark version as deprecated
+- Semantic versioning for patterns (major.minor.patch)
+- Version history with changelogs
+
+### Changed
+
+- Patterns module expanded from 6 to 9 tools
+
+## [2.10.0] - 2025-12-29
+
+### Added
+
+- **Consultation Module** (9 tools)
+  - `prepare_consultation_context` - Prepare session context for external LLM
+  - `capture_consultation_response` - Parse and persist consultation findings
+  - `validate_decision` - Prepare decision validation request
+  - `peer_review_session` - Comprehensive session review during finalization
+  - `peer_review_contract` - Contractor Model with acceptance criteria
+  - `validate_review_response` - Validate review against contract
+  - `seek_consensus` - Multi-model consensus queries
+  - `aggregate_consensus` - Synthesize multi-model responses
+  - `consult_llm_stream` - Direct LLM consultation with streaming
+- Fresh Eyes mode for multi-model validation
+- Cross-model consensus patterns
+
+### Changed
+
+- Added ADR-013: Consultation and Cross-Model Validation
+
+## [2.9.0] - 2025-12-28
+
+### Added
+
+- **Memory Bank Bridge** (Phase 4)
+  - Historical context seeding from Memory Bank
+  - Consultation response persistence to Memory Bank
+  - Cross-session learning via Vertex AI
+
+### Changed
+
+- Session coordination enhanced with Memory Bank integration
+
+## [2.8.0] - 2025-12-27
+
+### Added
+
+- **Graph Module** (3 tools)
+  - `add_note_relationship` - Link notes with typed relationships
+  - `query_temporal_graph` - Traverse causal chains, evolution paths
+  - `link_sessions` - Create relationships between sessions
+- Relationship types: preceded_by, caused_by, validated_by, superseded_by, related_to
+- Graph queries: causal_chain, temporal_sequence, relevant_adjacency, evolution_path, validation_chain
+
+### Changed
+
+- Added ADR-014: Temporal Graph Relationships
+
+## [2.7.0] - 2025-12-26
+
+### Added
+
+- **Memory Bank Module** (2 tools)
+  - `memory_bank_search` - Semantic search across project memories
+  - `memory_bank_list` - List all memories for a project
+- Vertex AI integration for long-term memory
+
+### Changed
+
+- Added ADR-012: Memory Bank Architecture
+
+## [2.6.0] - 2025-12-25
+
+### Added
+
+- **Research Module** (5 tools)
+  - `generateResearchPlan` - Create execution plan for user review
+  - `approveResearchPlan` - Approve/modify/cancel pending plan
+  - `trackResearchProgress` - Track execution status
+  - `listResearchSessions` - View all research sessions (inbox pattern)
+  - `suggestModelForResearch` - Recommend optimal model by domain
+- Plan-review-execute workflow
+- AA-Omniscience benchmark integration for model suggestions
+
+### Changed
+
+- Added ADR-016: Research Workflow Management
+
+## [2.5.0] - 2025-12-24
+
+### Added
+
+- **Approval Fingerprinting** (1 tool)
+  - `trackApprovalFingerprint` - Reduce approval fatigue via SHA-256 fingerprints
+- Session-scoped approval tracking
+- Operation type support: shell, edit, create, delete
+
+### Changed
+
+- Added ADR-015: Approval Fingerprinting
+
+## [2.4.0] - 2025-12-23
+
+### Added
+
+- **Constraint Tracking** (4 tools in Session module)
+  - `track_constraint` - Track constraints with keyword extraction
+  - `get_constraints` - Get all active constraints
+  - `lift_constraint` - Deactivate a constraint
+  - `check_constraint_violation` - Check if action violates constraints
+- Hybrid RRF scoring for constraint violation detection
+
+## [2.3.0] - 2025-12-22
+
+### Added
+
+- **Health Module** (2 tools)
+  - `getContextHealth` - Monitor context window health
+  - `getQdrantHealth` - Check Qdrant connection status
+- QdrantConnectionManager with auto-reconnection
+
+### Changed
+
+- Improved error handling and connection resilience
+
+## [2.2.0] - 2025-12-21
+
+### Added
+
+- **Stuck Module** (2 tools)
+  - `check_stuck_pattern` - Detect repeated blockers, error loops
+  - `get_recovery_suggestions` - Generate ranked recovery suggestions
+- 10-minute cooldown to prevent alert spam
+
+## [2.1.0] - 2025-12-20
+
+### Added
+
+- **Session Module** (7 core tools)
+  - `start_session_coordination` - Initialize Qdrant session memory
+  - `save_session_note` - Save decision/hypothesis/blocker/finding
+  - `session_search` - Fast semantic search (50-200ms)
+  - `check_duplicate_work` - Detect duplicate implementations
+  - `get_session_stats` - Get session statistics
+  - `extract_session_memories` - Extract learnings for finalization
+  - `finalize_session_coordination` - Cleanup ephemeral memory
+- Qdrant Cloud integration for real-time session coordination
+- OpenAI embeddings for semantic search
+
+## [2.0.0] - 2025-12-19
+
+### Added
+
+- **Search Module** (3 tools)
+  - `semanticSearch` - Query artifacts using Google File Search (99.1% token reduction)
+  - `indexSession` - Index session to File Search store
+  - `getSearchStats` - Get indexing statistics and costs
+- Google Gemini integration
+
+### Breaking Changes
+
+- Restructured tool organization into modules
+
+## [1.5.0] - 2025-12-15
+
+### Added
+
+- **Metrics Module** (2 tools)
+  - `getCompressionRatio` - Calculate token compression
+  - `getPatternReuse` - Track pattern reuse statistics
+- Cost tracking dashboard
+
+## [1.4.0] - 2025-12-10
+
+### Added
+
+- **Memory Module** (3 tools)
+  - `addNote` - Track decisions/hypotheses/blockers
+  - `getDecisions` - Retrieve session decisions
+  - `getHypotheses` - Retrieve session hypotheses
+
+## [1.3.0] - 2025-12-05
+
+### Added
+
+- **Artifacts Module** (4 tools)
+  - `searchArtifacts` - Search finalization packs by keyword
+  - `loadSession` - Load session context with query filtering
+  - `getSessionCode` - Extract executable code from sessions
+  - `extractSessionContext` - Extract compressed context for handoffs
+
+## [1.2.0] - 2025-11-25
+
+### Added
+
+- **Patterns Module** (6 tools)
+  - `searchPatterns` - Search by keyword/category
+  - `loadSkill` - Load documentation and interface
+  - `executeSkill` - Execute with parameters
+  - `findPatterns` - Intelligent search with Qdrant/Google FS
+  - `trackPatternUsage` - Track usage for auto-promotion
+  - `indexPatternLibrary` - Index patterns to vector stores
+- Progressive skill loading (98.7% token reduction)
+- Three-file skill pattern (SKILL.md + .md + .ts)
+
+## [1.1.0] - 2025-11-15
+
+### Added
+
+- TypeScript configuration with strict mode
+- Vitest test infrastructure
+- Token estimation utilities
+
+## [1.0.0] - 2025-11-05
+
+### Added
+
+- Initial MCP server implementation
+- stdio transport for Claude Code integration
+- Basic project structure
+
+---
+
+[2.20.0]: https://github.com/Szermer/centient/compare/v2.19.0...v2.20.0
+[2.19.0]: https://github.com/Szermer/centient/compare/v2.14.0...v2.19.0
+[2.14.0]: https://github.com/Szermer/centient/compare/v2.13.0...v2.14.0
+[2.13.0]: https://github.com/Szermer/centient/compare/v2.12.0...v2.13.0
+[2.12.0]: https://github.com/Szermer/centient/compare/v2.11.0...v2.12.0
+[2.11.0]: https://github.com/Szermer/centient/compare/v2.10.0...v2.11.0
+[2.10.0]: https://github.com/Szermer/centient/compare/v2.9.0...v2.10.0
+[2.9.0]: https://github.com/Szermer/centient/compare/v2.8.0...v2.9.0
+[2.8.0]: https://github.com/Szermer/centient/compare/v2.7.0...v2.8.0
+[2.7.0]: https://github.com/Szermer/centient/compare/v2.6.0...v2.7.0
+[2.6.0]: https://github.com/Szermer/centient/compare/v2.5.0...v2.6.0
+[2.5.0]: https://github.com/Szermer/centient/compare/v2.4.0...v2.5.0
+[2.4.0]: https://github.com/Szermer/centient/compare/v2.3.0...v2.4.0
+[2.3.0]: https://github.com/Szermer/centient/compare/v2.2.0...v2.3.0
+[2.2.0]: https://github.com/Szermer/centient/compare/v2.1.0...v2.2.0
+[2.1.0]: https://github.com/Szermer/centient/compare/v2.0.0...v2.1.0
+[2.0.0]: https://github.com/Szermer/centient/compare/v1.5.0...v2.0.0
+[1.5.0]: https://github.com/Szermer/centient/compare/v1.4.0...v1.5.0
+[1.4.0]: https://github.com/Szermer/centient/compare/v1.3.0...v1.4.0
+[1.3.0]: https://github.com/Szermer/centient/compare/v1.2.0...v1.3.0
+[1.2.0]: https://github.com/Szermer/centient/compare/v1.1.0...v1.2.0
+[1.1.0]: https://github.com/Szermer/centient/compare/v1.0.0...v1.1.0
+[1.0.0]: https://github.com/Szermer/centient/releases/tag/v1.0.0