{"$schema":"https://wellknown.network/schemas/agent-record-v1.json","schemaVersion":"1","id":"ag_p835573cac86","handle":"memocean-mcp","url":"https://wellknown.network/agents/memocean-mcp","links":{"self":"https://wellknown.network/agents/memocean-mcp/record.json","html":"https://wellknown.network/agents/memocean-mcp","markdown":"https://wellknown.network/agents/memocean-mcp/record.md","api":"https://wellknown.network/api/v1/agents/memocean-mcp","status":"https://wellknown.network/api/v1/agents/memocean-mcp/status","claim":"https://wellknown.network/agents/memocean-mcp/claim","claimApi":"https://wellknown.network/api/v1/claims","claimDescriptor":"https://wellknown.network/agents/memocean-mcp/claim.json","badge":"https://wellknown.network/agents/memocean-mcp/badge.svg","openapi":"https://wellknown.network/openapi.json","history":"https://wellknown.network/api/v1/agents/memocean-mcp/history","tools":"https://wellknown.network/api/v1/agents/memocean-mcp/tools"},"ard":{"identifier":"urn:air::server:memocean-mcp","type":"application/mcp-server-card+json"},"kind":"mcp_server","declared":{"name":"memocean-mcp","summary":"Local-first AI memory MCP server for Obsidian vaults. CJK-optimized.","description":"# memocean-mcp — Local-first AI Memory MCP Server\n\n![MemOcean](assets/memocean-banner.jpg)\n\n[![License: MIT](https://img.shields.io/badge/License-MIT-blue.svg)](LICENSE)\n[![Python 3.11+](https://img.shields.io/badge/Python-3.11%2B-blue.svg)](https://www.python.org/)\n\nA local MCP server for AI agents to search and manage knowledge stored in Obsidian vaults.\nBuilt for CJK (Chinese/Japanese/Korean) developers — 94.4% Hit@5 on Chinese queries with zero AI components required.\n\n**Core features:**\n- BM25/INSTR hybrid search — CJK-optimized, pure SQLite, no embeddings needed\n- CLSC sonar compression — 92.5% token reduction (13x compression) on Obsidian notes\n- Temporal knowledge graph — entity-relationship store with non-destructive invalidation\n- Cross-agent memory sharing — multiple agents share one `memory.db`\n- FATQ task queue — File-Atomic Task Queue for agent coordination\n\n---\n\n## Quick Start\n\n```bash\npip install memocean-mcp\n```\n\nAdd to Claude Desktop / Claude Code `.mcp.json`:\n\n```json\n{\n  \"mcpServers\": {\n    \"memocean\": {\n      \"command\": \"memocean-mcp\",\n      \"env\": {\n        \"MEMOCEAN_VAULT_ROOT\": \"/path/to/your/obsidian/vault\"\n      }\n    }\n  }\n}\n```\n\nOr register with Claude Code CLI:\n\n```bash\nMEMOCEAN_VAULT_ROOT=/path/to/vault claude mcp add memocean memocean-mcp\n```\n\n---\n\n## Environment Variables\n\n| Variable | Default | Description |\n|---|---|---|\n| `MEMOCEAN_VAULT_ROOT` | `~/Documents/Obsidian Vault` | Root of your Obsidian vault |\n| `MEMOCEAN_DATA_DIR` | `~/.memocean` | Data directory (databases, task queue) |\n| `MEMOCEAN_VAULT_PATH` | `MEMOCEAN_VAULT_ROOT/Ocean` | Ocean subdirectory for full-text search |\n| `MEMOCEAN_SKILLS_DIR` | `MEMOCEAN_VAULT_ROOT/Ocean/Pearl/skills` | Skills markdown directory |\n| `MEMOCEAN_USE_GBRAIN` | `false` | Enable GBrain hybrid search delegate |\n| `KNN_ENABLED` | `false` | Enable BGE-m3 KNN vector search |\n| `ENABLE_QUERY_EXPANSION` | unset | Enable Haiku query expansion (requires `ANTHROPIC_API_KEY`) |\n| `ENABLE_HAIKU_RERANKE…","publisher":null,"homepage":"https://github.com/ChannelLabAI/memocean","repository":"https://github.com/ChannelLabAI/memocean","version":"0.5.0","license":"MIT","protocols":["mcp"],"tags":["mcp","obsidian","memory","ai","claude","cjk","chinese"],"pricing":null,"endpoints":[{"url":"pypi:memocean-mcp","type":"package_pypi","auth":null,"probeable":false}],"skills":null,"tools":null,"extra":null,"attribution":{"kind":"pypi","name":"pypi","license":"pypi","repoUrl":"pypi","summary":"pypi","version":"pypi","description":"pypi","homepageUrl":"pypi"}},"derived":{"capabilities":[{"slug":"data.database","name":"Databases","confidence":1,"provenance":"derived"},{"slug":"productivity.notes","name":"Notes & Docs","confidence":1,"provenance":"declared"},{"slug":"knowledge.memory","name":"Agent Memory","confidence":1,"provenance":"declared"},{"slug":"knowledge.knowledge-graph","name":"Knowledge Graphs","confidence":0.836,"provenance":"derived"},{"slug":"data.vector-search","name":"Vector Search","confidence":0.825,"provenance":"derived"}],"categories":["data","knowledge","productivity"],"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":"memocean-mcp","registry":"pypi","observedAt":"2026-10-28T15:26:03.705Z","publishedAt":"2026-04-25T08:44:09.006788Z","latestVersion":"0.5.0"},"toolSurface":null,"endpointFacts":[]},"verification":{"claimed":false,"claimedAt":null,"proofs":[]},"provenance":{"sources":[{"source":"pypi","key":"memocean-mcp","url":"https://pypi.org/project/memocean-mcp/","firstSeenAt":"2026-09-10T05:22:47.623Z","fetchedAt":"2026-10-28T15:24:34.307Z","normalizedAt":"2026-10-28T15:24:34.307Z"}]},"firstSeenAt":"2026-09-10T05:22:47.623Z","updatedAt":"2026-10-28T15:26:03.705Z"}