TODO — PKM Bridge
Lists 28 open tasks for maintaining a personal knowledge management bridge between a custom app and MCP interfaces.
What this file does
Lists 28 open tasks for maintaining a personal knowledge management bridge between a custom app and MCP interfaces.
When to use it
- You maintain a PKM system with dual interfaces
- You need a structured backlog for tool and search quality improvements
- You plan to add embeddings or RAG to a note-taking setup
- You want to track self-improvement agent and infrastructure work items
Assumes this stack
TODO — PKM Bridge
Current and planned work items. See PROJECT.md for architecture overview.
Dual Interface Maintenance
- Keep system prompts aligned on core PKM behavior (search strategy, file safety, note creation)
- Test both interfaces when changing tools or backend logic
- Ensure new skills work correctly in both custom app and MCP contexts
- Monitor self-improvement agent changes for cross-interface impact
Tool & Search Quality
- Improve semantic search ranking and chunk quality
- Better handling of date-range queries across both org and Logseq
- Reduce unnecessary tool calls in common workflows
- Add tool usage analytics to identify patterns and failures
Embeddings / RAG
- Tune chunking strategy for better retrieval precision
- Add embedding coverage monitoring (% of notes indexed)
- Explore re-ranking retrieved chunks before injection
Self-Improvement Agent
- Review and consolidate accumulated skills for redundancy
- Add conversation quality metrics the agent can track
- Improve feedback signal detection from implicit user behavior
Frontend / UX
- Editor SPA: collaborative editing awareness (external file changes)
- Mobile responsiveness improvements
- Session management UX (list, rename, delete sessions)
Infrastructure
- Health check endpoint improvements (check both :8000 and :8001)
- Structured logging for better debugging
- Backup strategy for PostgreSQL data
- Monitor Docker resource usage
Future Ideas
- Calendar event creation from natural language
- Cross-note link suggestions based on embeddings
- Automated daily/weekly summaries
- Org-roam graph integration
- Push notifications for reminders
What's inside
7 sections with 28 checklist items across dual interface, tools, embeddings, agent, UX, infrastructure, and future ideas.
Change this for your project
- Replace
org and Logseqwith your own note-taking tools - Replace
:8000 and:8001with your service port numbers - Replace
PostgreSQLwith your database system if different
Where it goes
Reference documentation for a retrieval pipeline. Keep with the ingestion or retrieval code it describes.
Related Documents
SUMMARY
Proposes three on-prem AI architectures, modular, hybrid, and fully local RAG, with hardware specs and vendor lists.
Retrieval & Prompts
Explains how CharMemory's extraction prompt and Vector Storage settings determine memory retrieval quality in SillyTavern.
Advanced RAG Techniques
Surveys advanced RAG architectures, Agentic, Corrective, Self-RAG, Graph RAG, RAPTOR, Multimodal, Temporal, with design trade-offs and decision guidance.
RFC-BLite: High-Performance Embedded Document Database for .NET
Specifies an embedded document database for.NET with zero-allocation I/O, C-BSON format, and ACID transactions.