{"id":"lethain-library-mcp","name":"library-mcp","homepage":null,"repo_url":"https://github.com/lethain/library-mcp","category":"ai-ml","subcategories":[],"tags":["ai-ml","devtools","infrastructure","api-gateway","search"],"what_it_does":"Provides an MCP server (Python) to query local Markdown knowledge bases. It indexes folders of .md files with front-matter metadata (e.g., title/tags/url) and exposes MCP tools to retrieve content by tag/text/slug/date range, manage tags, and rebuild the index.","use_cases":["Local RAG over Markdown folders for blog/wiki-style content","Generating summaries or recommendations based on tags or time ranges","Interactive research workflows with an MCP-enabled client (e.g., Claude Desktop)","Searching a knowledge base for relevant posts/snippets by text or identifier"],"not_for":["Serving untrusted remote users over the public internet without additional hardening","Use as a secure multi-tenant document database","Use where strict auditability/compliance requirements require documented guarantees"],"best_when":"You run it locally (or in a trusted internal environment) to query your own Markdown corpus and want an agent to fetch relevant excerpts rather than sending entire documents to a model context window.","avoid_when":"You need robust network security controls (TLS termination, strong auth, tenant isolation) or documented SLAs/error-retry semantics; also avoid exposing it to untrusted input without sandboxing and validation.","alternatives":["RAG with a hosted vector database over your Markdown (e.g., embeddings + search)","Local search/indexing tools (ripgrep + custom scripts) combined with an LLM","Other MCP servers for file/content retrieval (if available)","Static-site index/search solutions (if your workflow is mostly lookup)"],"af_score":43.2,"security_score":28.0,"reliability_score":17.5,"package_type":"mcp_server","discovery_source":["github"],"priority":"high","status":"evaluated","version_evaluated":null,"last_evaluated":"2026-03-30T13:48:46.356418+00:00","interface":{"has_rest_api":false,"has_graphql":false,"has_grpc":false,"has_mcp_server":true,"mcp_server_url":null,"has_sdk":false,"sdk_languages":[],"openapi_spec_url":null,"webhooks":false},"auth":{"methods":["MCP transport configuration (implied by MCP server usage)"],"oauth":false,"scopes":false,"notes":"No authentication mechanism is described in the README. Since it’s intended for local hobby use, it likely relies on local access rather than auth controls."},"pricing":{"model":null,"free_tier_exists":false,"free_tier_limits":null,"paid_tiers":[],"requires_credit_card":false,"estimated_workload_costs":null,"notes":"Open-source/local tool; no pricing info provided."},"requirements":{"requires_signup":false,"requires_credit_card":false,"domain_verification":false,"data_residency":[],"compliance":[],"min_contract":null},"agent_readiness":{"af_score":43.2,"security_score":28.0,"reliability_score":17.5,"mcp_server_quality":60.0,"documentation_accuracy":50.0,"error_message_quality":0.0,"error_message_notes":null,"auth_complexity":95.0,"rate_limit_clarity":0.0,"tls_enforcement":10.0,"auth_strength":15.0,"scope_granularity":10.0,"dependency_hygiene":55.0,"secret_handling":60.0,"security_notes":"Intended for local use; authentication, authorization, and network security controls are not documented. It processes local Markdown files with front-matter; treat corpus as untrusted input if it can contain malicious content. No guarantees are stated about safe parsing, sandboxing, or avoiding path traversal—additional validation may be required if exposed beyond localhost.","uptime_documented":0.0,"version_stability":45.0,"breaking_changes_history":0.0,"error_recovery":25.0,"idempotency_support":"false","idempotency_notes":"A 'rebuild' tool exists, but idempotency and safe retry behavior are not documented.","pagination_style":"none","retry_guidance_documented":false,"known_agent_gotchas":["No documented rate limits or pagination/streaming contract for tool outputs; agents may need to cap requests (e.g., 'next 50 posts') themselves.","No documented error codes/structures; agent error recovery and retry strategies may require probing.","Since it indexes local folders, performance may degrade if corpus is large or rebuild is frequent."]}}