{"$schema":"https://wellknown.network/schemas/agent-record-v1.json","schemaVersion":"1","id":"ag_batu3cxgra9p","handle":"semantic-memory-mcp","url":"https://wellknown.network/agents/semantic-memory-mcp","links":{"self":"https://wellknown.network/agents/semantic-memory-mcp/record.json","html":"https://wellknown.network/agents/semantic-memory-mcp","markdown":"https://wellknown.network/agents/semantic-memory-mcp/record.md","api":"https://wellknown.network/api/v1/agents/semantic-memory-mcp","status":"https://wellknown.network/api/v1/agents/semantic-memory-mcp/status","claim":"https://wellknown.network/agents/semantic-memory-mcp/claim","claimApi":"https://wellknown.network/api/v1/claims","claimDescriptor":"https://wellknown.network/agents/semantic-memory-mcp/claim.json","badge":"https://wellknown.network/agents/semantic-memory-mcp/badge.svg","openapi":"https://wellknown.network/openapi.json","history":"https://wellknown.network/api/v1/agents/semantic-memory-mcp/history","tools":"https://wellknown.network/api/v1/agents/semantic-memory-mcp/tools"},"ard":{"identifier":"urn:air::server:semantic-memory-mcp","type":"application/mcp-server-card+json"},"kind":"mcp_server","declared":{"name":"semantic-memory-mcp","summary":"Single-file, self-hosted semantic memory MCP server for AI coding agents","description":"# Semantic Memory MCP\n\n**Persistent, semantic memory for AI coding agents — a single Python file, entirely local.**\n\n`semantic-memory-mcp` is a self-hosted [Model Context Protocol](https://modelcontextprotocol.io) server that gives AI coding agents a durable, searchable memory across sessions. It is built as one Python file, stores everything in a single local SQLite database, and needs no cloud service and no API key.\n\n## What it is\n\n- **Cross-session memory:** notes survive between sessions; the agent searches its own memory at the start of every task instead of relying on the user to re-explain context.\n- **Hybrid retrieval:** keyword full-text search (SQLite FTS5) and semantic vector search (sqlite-vec) are combined through Reciprocal Rank Fusion — a note can be found by exact wording *or* by meaning.\n- **Multilingual embeddings:** quantized ONNX model (`Xenova/multilingual-e5-small`, ~113 MB) runs locally on CPU. English and German work equally well; no OpenAI key required.\n- **Trivially auditable:** the whole server is a single readable file (`memory_server.py`, ~1,100 lines). There is no hidden state beyond the SQLite file.\n\n### Tools\n\n| Tool | Purpose |\n|------|---------|\n| `memory_store` | Save a note (with tags and source) |\n| `memory_search` | Hybrid search (semantic + full-text), optional token budget |\n| `memory_get` / `memory_recent` | Read one note by ID / list recent notes |\n| `memory_update` / `memory_delete` | Edit or remove a note |\n| `memory_status` | Database statistics |\n| `memory_zotero_sync` | Optional: back up memory content into a Zotero library |\n\n## Why\n\nAI coding agents forget everything between sessions. File-based memory (markdown indexes that get reloaded) is keyword-based: it has no semantic recall and no ranking — a note saved as *\"embedding model comparison\"* will not be found when the agent later asks *\"which embedding did we test?\"*.\n\nThis server was built as a working reference implementation and **measured, not assumed** — see…","publisher":{"name":"Michael Logies","url":null},"homepage":"https://github.com/Michael-Logies/semantic-memory-mcp","repository":"https://github.com/Michael-Logies/semantic-memory-mcp","version":"1.0.4","license":null,"protocols":["mcp"],"tags":["mcp"],"pricing":null,"endpoints":[{"url":"pypi:semantic-memory-mcp","type":"package_pypi","auth":null,"probeable":false}],"skills":null,"tools":null,"extra":null,"attribution":{"kind":"pypi","name":"pypi","repoUrl":"pypi","summary":"pypi","version":"pypi","description":"pypi","homepageUrl":"pypi","publisherName":"pypi"}},"derived":{"capabilities":[{"slug":"knowledge.memory","name":"Agent Memory","confidence":1,"provenance":"derived"},{"slug":"data.vector-search","name":"Vector Search","confidence":0.825,"provenance":"derived"},{"slug":"data.database","name":"Databases","confidence":0.802,"provenance":"derived"}],"categories":["data","knowledge"],"language":"en"},"observed":{"status":"unknown","statusReason":"Distributed as a package to run locally; no network endpoint to check.","lastOkAt":null,"lastProbedAt":null,"statusComputedAt":null,"reliability30d":null,"latestObservations":[],"tools":null,"package":{"name":"semantic-memory-mcp","registry":"pypi","observedAt":"2026-11-05T22:26:31.359Z","publishedAt":"2026-10-31T10:27:39.883659Z","latestVersion":"1.0.4"},"toolSurface":null,"endpointFacts":[]},"verification":{"claimed":false,"claimedAt":null,"proofs":[]},"provenance":{"sources":[{"source":"pypi","key":"semantic-memory-mcp","url":"https://pypi.org/project/semantic-memory-mcp/","firstSeenAt":"2026-10-12T14:25:46.025Z","fetchedAt":"2026-11-05T22:25:29.732Z","normalizedAt":"2026-11-05T22:25:29.732Z"}]},"firstSeenAt":"2026-10-12T14:25:46.025Z","updatedAt":"2026-11-05T22:26:31.359Z"}