jonhardwick-spec/specmem
SpecMem - TRUE Semantic Memory for Claude Code - Runs locally - Remembers your code base - Per-project isolation, semantic search, team coordination. https://justcalljon.pro
Leverages PostgreSQL with pgvector for semantic code indexing and ONNX-quantized embeddings running in an air-gapped Docker container, ensuring code never leaves the local machine. Exposes 74+ MCP tools across 14 language extractors, with a three-layer transport stack (proxy/bootstrap/MCP server) and PostgreSQL message bus enabling unlimited multi-agent swarm coordination. Includes automatic token compaction (~60% reduction), HNSW indexes for fast retrieval, and 10 Claude hooks for context injection and file claim enforcement across isolated project schemas.
Stars
19
Forks
2
Language
JavaScript
License
—
Category
Last pushed
Mar 10, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/mcp/jonhardwick-spec/specmem"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
patravishek/memex
Retains Claudes session
sgroy10/speclock
AI Constraint Engine by Sandeep Roy — stops AI from breaking what you locked. 100/100 on...
Intina47/context-sync
Local persistent memory store for LLM applications including continue.dev, cursor, claude...
amanhij/Zikkaron
Biologically-inspired persistent memory engine for Claude Code. 26 cognitive subsystems,...
roampal-ai/roampal-core
Outcome-based persistent memory MCP server for Claude Code and OpenCode. Good advice promoted,...