{"$schema":"https://wellknown.network/schemas/agent-record-v1.json","schemaVersion":"1","id":"ag_4ntwtgwxsajt","handle":"akasha-mcp","url":"https://wellknown.network/agents/akasha-mcp","links":{"self":"https://wellknown.network/agents/akasha-mcp/record.json","html":"https://wellknown.network/agents/akasha-mcp","markdown":"https://wellknown.network/agents/akasha-mcp/record.md","api":"https://wellknown.network/api/v1/agents/akasha-mcp","status":"https://wellknown.network/api/v1/agents/akasha-mcp/status","claim":"https://wellknown.network/agents/akasha-mcp/claim","claimApi":"https://wellknown.network/api/v1/claims","claimDescriptor":"https://wellknown.network/agents/akasha-mcp/claim.json","badge":"https://wellknown.network/agents/akasha-mcp/badge.svg","openapi":"https://wellknown.network/openapi.json","history":"https://wellknown.network/api/v1/agents/akasha-mcp/history","tools":"https://wellknown.network/api/v1/agents/akasha-mcp/tools"},"ard":{"identifier":"urn:air::server:akasha-mcp","type":"application/mcp-server-card+json"},"kind":"mcp_server","declared":{"name":"akasha-mcp","summary":"A local knowledge base for your coding agents, served over MCP.","description":"# akasha\n\nA local knowledge base that gives coding agents a memory, served over MCP.\n\n## The problem\n\nEvery agent session starts from zero. Yesterday an agent spent an hour working out why a\ntest fails one run in twenty: two tests share a fixture directory and the cleanup of one\nraces the setup of the other. It fixed the test and the session ended. The reasoning\nlived in a context window, and the window is gone. Today another agent meets the same\nflake, reads the same files and derives the same answer, spending the same time and\ncontext to get there.\n\nWhat you already know is not missing, it is scattered. Some of it sits in Claude Code\nproject memory, some in Serena memories, some in a notes folder, some in markdown files\ninside repos. Each agent writes to its own corner, and none of them searches the others.\nPasting notes into prompts does not scale: context is the scarce resource, and every\npasted page crowds out the work. Conventions (\"migrations are reviewed before merge\",\n\"never mock the database in integration tests\") get retyped at the start of session\nafter session.\n\n## How akasha fixes it\n\nakasha builds one local index over the markdown you already have. It is read-only by\ndefault and copies nothing: your files stay the source of truth, and the SQLite index\nderived from them can be deleted and rebuilt at any time.\n\nAgents search it through MCP before they start investigating, and write what they find\nback as ordinary markdown with `knowledge_write` and `knowledge_append`. The next\nsession starts where the last one ended. Search is keyword (BM25) and, if enabled, dense\nretrieval, fused by reciprocal rank. Results are sized for a context window: a handful of\ncapped hits, with `knowledge_get` to page into a document only when one is worth reading.\n\nConventions are documents of `kind=convention`. They are delivered automatically at\nsession start and are read-only to agents; only the CLI can create or change one. Hooks\nkeep the index fresh without anyone thinki…","publisher":{"name":"Ahmad Nur Salim","url":null},"homepage":"https://github.com/anuress/akasha","repository":"https://github.com/anuress/akasha","version":"0.0.6","license":null,"protocols":["mcp"],"tags":["claude-code","coding-agents","knowledge-base","mcp"],"pricing":null,"endpoints":[{"url":"pypi:akasha-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":"code.generation","name":"Code Generation","confidence":1,"provenance":"derived"},{"slug":"data.database","name":"Databases","confidence":0.802,"provenance":"derived"},{"slug":"dev.filesystem","name":"Filesystem","confidence":0.802,"provenance":"derived"}],"categories":["code","data","dev"],"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":"akasha-mcp","registry":"pypi","observedAt":"2026-11-08T22:25:16.345Z","publishedAt":"2026-11-07T07:25:37.394445Z","latestVersion":"0.0.6"},"toolSurface":null,"endpointFacts":[]},"verification":{"claimed":false,"claimedAt":null,"proofs":[]},"provenance":{"sources":[{"source":"pypi","key":"akasha-mcp","url":"https://pypi.org/project/akasha-mcp/","firstSeenAt":"2026-10-09T13:27:29.985Z","fetchedAt":"2026-11-08T22:24:26.702Z","normalizedAt":"2026-11-08T22:24:26.702Z"}]},"firstSeenAt":"2026-10-09T13:27:29.985Z","updatedAt":"2026-11-08T22:25:16.345Z"}