Skip to main content
Glama
IMPLEMENTATION_PLANS.mdβ€’8.28 kB
# Langfuse MCP Server Implementation Plans This document tracks the implementation status and future roadmap for the Langfuse MCP Server, helping prioritize development efforts and track progress toward complete API coverage. ## πŸ“Š Current Status (v1.3.0) **Total Implemented Tools: 28** - βœ… Core Analytics Tools: 6 - βœ… Extended Analytics Tools: 6 - βœ… System & Management Tools: 6 - βœ… Dataset Management Tools: 7 ⭐ **COMPLETED in v1.2.0** - βœ… Comment Management Tools: 3 ⭐ **COMPLETED in v1.3.0** **API Coverage: ~50% of full Langfuse API** ## 🎯 Remaining Implementation Phases ### πŸ”΄ **Phase 1: Core Collaboration (v1.3.0)** ⭐ **COMPLETED** **Priority:** ⭐⭐⭐⭐⭐ **HIGHEST** **Timeline:** 2-3 weeks **Business Impact:** Critical for team workflows #### **Comments System (3 tools)** βœ… **COMPLETED** - βœ… `create_comment` - Add comments to traces/observations/sessions/prompts - **API:** `POST /api/public/comments` - **Params:** objectType, objectId, content, authorUserId, projectId - βœ… `list_comments` - Get comments with filtering - **API:** `GET /api/public/comments` - **Filters:** objectType, objectId, authorUserId, pagination - βœ… `get_comment` - Get individual comment details - **API:** `GET /api/public/comments/{commentId}` #### **Session Management (2 tools)** ❌ **NOT AVAILABLE** *Analysis revealed that Langfuse API does not include dedicated session endpoints* - ❌ `list_sessions` - API endpoint does not exist - ❌ `get_session` - API endpoint does not exist - πŸ“ **Note:** Session data is handled through trace/observation entities **Phase 1 Total: 3 new tools (25 β†’ 28 total)** --- ### 🟑 **Phase 2: Advanced Workflows (v1.4.0)** **Priority:** ⭐⭐⭐⭐ **HIGH** **Timeline:** 3-4 weeks **Business Impact:** Important for ML evaluation workflows #### **Annotation Queues (9 tools)** - [ ] `create_annotation_queue` - Create annotation workflows - **API:** `POST /api/public/annotation-queues` - [ ] `list_annotation_queues` - Manage existing queues - **API:** `GET /api/public/annotation-queues` - [ ] `get_annotation_queue` - Get queue details - **API:** `GET /api/public/annotation-queues/{queueId}` - [ ] `create_queue_item` - Add items for annotation - **API:** `POST /api/public/annotation-queues/{queueId}/items` - [ ] `list_queue_items` - Get items from queues - **API:** `GET /api/public/annotation-queues/{queueId}/items` - [ ] `get_queue_item` - Get specific queue item - **API:** `GET /api/public/annotation-queues/{queueId}/items/{itemId}` - [ ] `update_queue_item` - Update annotation status - **API:** `PATCH /api/public/annotation-queues/{queueId}/items/{itemId}` - [ ] `delete_queue_item` - Remove items - **API:** `DELETE /api/public/annotation-queues/{queueId}/items/{itemId}` - [ ] `manage_queue_assignments` - Assign queues to team members - **APIs:** `POST/DELETE /api/public/annotation-queues/{queueId}/assignments` #### **Score Management (6 tools)** - [ ] `create_score` - Add scores to traces/observations/sessions - **API:** `POST /api/public/scores` - [ ] `list_scores` - Get scores with filtering - **API:** `GET /api/public/v2/scores` - [ ] `get_score` - Get individual score - **API:** `GET /api/public/v2/scores/{scoreId}` - [ ] `delete_score` - Remove scores - **API:** `DELETE /api/public/scores/{scoreId}` - [ ] `create_score_config` - Define score structures - **API:** `POST /api/public/score-configs` - [ ] `list_score_configs` - Manage score configurations - **API:** `GET /api/public/score-configs` **Phase 2 Total: 15 new tools (30 β†’ 45 total)** --- ### ⚑ **Phase 3: Complete CRUD (v1.5.0)** **Priority:** ⭐⭐⭐ **MEDIUM** **Timeline:** 2-3 weeks **Business Impact:** Enhanced power-user features #### **Model CRUD Enhancement (2 tools)** *Current: Read-only operations* - [ ] `create_model` - Register custom models - **API:** `POST /api/public/models` - [ ] `delete_model` - Remove models - **API:** `DELETE /api/public/models/{id}` #### **Prompt CRUD Enhancement (2 tools)** *Current: Read-only operations* - [ ] `create_prompt` - Create new prompt versions - **API:** `POST /api/public/v2/prompts` - [ ] `update_prompt_label` - Manage prompt versioning/labels - **API:** `PATCH /api/public/v2/prompts/{name}/labels/{label}` #### **LLM Provider Management (2 tools)** - [ ] `list_llm_connections` - List configured providers - **API:** `GET /api/public/llm-connections` - [ ] `upsert_llm_connection` - Configure LLM provider connections - **API:** `PUT /api/public/llm-connections` #### **Advanced Operations (2 tools)** - [ ] `delete_trace` - Remove traces - **API:** `DELETE /api/public/traces/{traceId}` - [ ] `list_dataset_runs` - Get evaluation run history - **API:** `GET /api/public/datasets/{datasetName}/runs` **Phase 3 Total: 8 new tools (45 β†’ 53 total)** --- ### 🏒 **Phase 4: Enterprise Features (v2.0.0)** **Priority:** ⭐⭐ **LOWER** **Timeline:** 3-4 weeks **Business Impact:** Enterprise and integration features #### **Media & File Management (3 tools)** - [ ] `create_media_upload_url` - Get presigned upload URLs - **API:** `POST /api/public/media` - [ ] `get_media` - Retrieve media files - **API:** `GET /api/public/media/{mediaId}` - [ ] `update_media` - Update media records - **API:** `PATCH /api/public/media/{mediaId}` #### **Data Integration (3 tools)** - [ ] `batch_ingest` - Bulk data ingestion - **API:** `POST /api/public/ingestion` - [ ] `create_blob_storage_integration` - Configure blob storage - **API:** `PUT /api/public/integrations/blob-storage` - [ ] `list_blob_storage_integrations` - Manage storage configs - **API:** `GET /api/public/integrations/blob-storage` #### **Organization Management (6+ tools)** *Requires organization-scoped API keys* - [ ] `list_organization_projects` - Organization-level project listing - [ ] `create_project` - Project CRUD operations - [ ] `manage_api_keys` - API key lifecycle management - [ ] `list_organization_users` - Team management (SCIM) - [ ] `create_organization_user` - Add team members (SCIM) - [ ] `delete_organization_user` - Remove team members (SCIM) **Phase 4 Total: 12+ new tools (53 β†’ 65+ total)** --- ## πŸ“ˆ **Progress Tracking** ### **Completed Phases** - [x] **v1.0.0** - Initial implementation (18 tools) - [x] **v1.1.0** - Enhanced testing and documentation - [x] **v1.2.0** - Dataset Management (7 new tools) βœ… **COMPLETED** - [x] **v1.3.0** - Comment Management (3 new tools) βœ… **COMPLETED** ### **Upcoming Phases** - [ ] **v1.4.0** - Advanced Workflows (15 new tools) - [ ] **v1.5.0** - Complete CRUD (8 new tools) - [ ] **v2.0.0** - Enterprise Features (12+ new tools) ### **Target Metrics** - **Current API Coverage:** ~50% (28/55+ possible tools) - **v1.4.0 Target:** ~80% (43/55+ tools) - **v1.5.0 Target:** ~95% (51/55+ tools) - **v2.0.0 Target:** ~98% (63+/65+ tools) ## 🎯 **Implementation Strategy** ### **Phase Selection Criteria** 1. **Business Impact** - Features that enable key ML workflows 2. **User Demand** - Most requested functionality 3. **Implementation Complexity** - Leverage existing patterns 4. **API Dependencies** - Some features require organization-scoped keys ### **Technical Considerations** - **Consistent Patterns** - Follow established tool/client/test structure - **Error Handling** - Maintain robust error management - **Testing** - Comprehensive test coverage for each new tool - **Documentation** - Update docs with each release ### **Next Recommended Phase** **Phase 2 (Annotation Queues & Score Management)** because: - ⭐ **High business value for ML evaluation workflows** - πŸ”§ **Builds on existing dataset functionality** - πŸ‘₯ **Enables advanced team collaboration** - πŸ“Š **Critical for ML model evaluation and improvement** ## πŸ“ **Notes** - This plan covers ~95% of the public Langfuse API surface - Some enterprise features require organization-scoped API keys - Implementation order can be adjusted based on user feedback - Each phase includes comprehensive testing and documentation updates - Focus on maintaining the high-quality patterns established in v1.2.0 --- **Last Updated:** November 9, 2024 (v1.3.0 completion) **Next Target:** Phase 2 - Advanced Workflows (v1.4.0)

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/therealsachin/langfuse-mcp-server'

If you have feedback or need assistance with the MCP directory API, please join our Discord server