Show HN: DocuMemory – Persistent memory and context for AI coding tools

DocuMemory launched on Show HN as a persistent-memory tool that distills raw AI chat sessions into reusable context blocks for AI coding clients including Claude Code, ChatGPT and Cursor, connecting via a Model Context Protocol server at documemory.dev/api/mcp. The tool stores memory in the user's own PostgreSQL with pgvector HNSW indexing and combines semantic and full-text search through reciprocal rank fusion, with a local-first, self-hosted option. Pricing starts with 15 free distillations, then US $12/month or US $120 billed annually (US $10/month), and US $15 per seat per month for teams with a 2-seat minimum and no free trial.

BUILT FOR YOUR NEXT SESSION The Living Memory Engine for AI-First Engineers Passively distil raw AI chat sessions into permanent architecture decisions and reusable, prompt-injectable context blocks. 01 / THE MEMORY PIPELINE From conversation to compounding knowledge. Keep the insight. Lose the noise. Give every new session a head start. Ask your AI client to save the session. Capture the reasoning in the same place you do the work. save chat as doc { transcript, project id } 02 / SIGNAL OVER NOISE A long conversation. A lasting memory. Your next AI session doesn't need every debugging attempt. It needs what worked, why it worked, and the constraints worth remembering. --- doc type: MEMORY project: Platform tags: postgres, pgvector, search --- Search architecture Key facts - PostgreSQL + pgvector with HNSW indexing. - Semantic and full-text arms run independently. - Reciprocal rank fusion combines ranked results. - Every query is scoped to the authenticated user. ai context snippet Use PostgreSQL + pgvector with HNSW indexing. Fuse semantic and full-text search with RRF. Keep results scoped to the authenticated user. 03 / ENGINEERED FOR CONTINUITY A second brain for your engineering stack. Passive Chat-to-Doc Distillation Stay in Claude Code, ChatGPT, or Cursor. One capture turns your conversation into documentation that lasts. Model Context Protocol MCP Native Capture, search, and recall from your AI client. Your engineering knowledge becomes a tool your model can use. Subtractive Noise Reduction Zero-Fluff Code Keep the final implementation and the reasoning behind it. Leave the dead ends behind. Hybrid Vector & Full-Text Search RRF Semantic understanding meets exact identifiers. Reciprocal rank fusion brings the right context to the surface. Local-First & Self-Hosted Privacy Your infrastructure, your Postgres, your AI keys. Self-host with a local model to keep your knowledge on your machine. Prompt Real-Estate Optimization Compact context snippets give your next session the facts it needs without the transcript it doesn't. PRICING Try it first. Subscribe when you're ready. 15 free distillations. No credit card required. Free trial US $0 15 uses, once per account Explore your workspace and connect your AI client. Distillation and Enhance with AI share the free allowance. Try for free https://www.documemory.dev/signup monthly US $12/month US $12 billed monthly Continue after your free uses with documents, AI integrations, and private uploads. Cancel renewal from Settings. Signup now https://www.documemory.dev/signup annual Save US $24/year US $10/month US $120 billed annually Continue after your free uses with documents, AI integrations, and private uploads. Cancel renewal from Settings. Signup now https://www.documemory.dev/signup Team US $15/seat per month, or US $150 per seat, billed annually · 2-seat minimum One shared documentation repository for your team. Every member gets a seat; share projects with everyone or restrict them to specific people. No free trial: billing starts when you set up the organization. Start a team https://www.documemory.dev/signup?next=%2Forganizations%2Fnew 04 / CONNECT YOUR CLIENT One connection. Every decision, remembered. Create an access token in Settings, add your MCP connection, and ask your client to save a conversation. { "mcpServers": { "documemory": { "url": "https://documemory.dev/api/mcp", "headers": { "Authorization": "Bearer YOUR ACCESS TOKEN" } } } }