{"id":"biocontext-ai-skill-to-mcp","name":"skill-to-mcp","homepage":"https://biocontext.ai/registry/biocontext-ai/skill-to-mcp","repo_url":"https://github.com/biocontext-ai/skill-to-mcp","category":"devtools","subcategories":[],"tags":["mcp","fastmcp","skills","claude-skills","retrieval","filesystem","agent-tools","biocontextai"],"what_it_does":"Provides an MCP server that converts a collection of Claude Skills (organized as skill directories with SKILL.md YAML frontmatter) into MCP-accessible resources. It auto-discovers skills and exposes tools to list skills, fetch skill details (SKILL.md content plus file listings), and read related files within a skill directory with traversal protection.","use_cases":["Expose a local collection of Claude Skills to MCP-capable LLM apps","Programmatically browse skill metadata (name/description) for retrieval or planning","Fetch skill documentation/content and supporting files (scripts/references/examples) for downstream agents","Maintain multiple skill collections via different SKILLS_DIR configurations (dev/prod or domain-specific)"],"not_for":["Serving remote or multi-tenant untrusted users (package appears oriented to local filesystem skills)","Providing a general-purpose file server over arbitrary paths (only skill-directory scoped access is intended)","Production SaaS deployments that require built-in user authentication/authorization"],"best_when":"You have a curated set of skill directories on disk and want an MCP client/agent to retrieve skill content and associated files safely and consistently.","avoid_when":"You need strong user-level access control, rate limiting guarantees, or audited multi-tenant security controls out of the box.","alternatives":["Write a custom MCP server tailored to your skill schema and file layout","Use an MCP directory/document connector that indexes markdown/docs and serves them without the Claude Skills structure","Convert skills into a knowledge base (e.g., vector DB + document store) and expose via a retrieval-focused MCP server"],"af_score":63.5,"security_score":51.2,"reliability_score":36.2,"package_type":"mcp_server","discovery_source":["github"],"priority":"high","status":"evaluated","version_evaluated":null,"last_evaluated":"2026-03-30T15:33:13.480843+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":["none (based on provided documentation; authentication/authorization not described)"],"oauth":false,"scopes":false,"notes":"The README describes local configuration for SKILLS_DIR and MCP transport options but does not describe any authentication mechanism."},"pricing":{"model":null,"free_tier_exists":false,"free_tier_limits":null,"paid_tiers":[],"requires_credit_card":false,"estimated_workload_costs":null,"notes":"Open-source package; 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":63.5,"security_score":51.2,"reliability_score":36.2,"mcp_server_quality":70.0,"documentation_accuracy":75.0,"error_message_quality":0.0,"error_message_notes":null,"auth_complexity":95.0,"rate_limit_clarity":5.0,"tls_enforcement":35.0,"auth_strength":15.0,"scope_granularity":80.0,"dependency_hygiene":70.0,"secret_handling":70.0,"security_notes":"README states 'directory traversal protection' for get_skill_related_file, which is a strong scoped-access control. However, authentication/authorization is not described, and rate limiting is not mentioned. TLS/HTTPS requirements for HTTP transport are not documented here. Secrets are not discussed; likely low risk for secrets because the primary inputs are local paths and skill files, but details are not provided.","uptime_documented":0.0,"version_stability":55.0,"breaking_changes_history":40.0,"error_recovery":50.0,"idempotency_support":"true","idempotency_notes":"Read-only tools (list skills / get details / read related file) are inherently safe to retry; idempotency is likely good for typical use.","pagination_style":"none","retry_guidance_documented":false,"known_agent_gotchas":["Requires a correctly structured skills directory containing SKILL.md files; missing/malformed frontmatter may affect discovery/metadata extraction.","File access is scoped to the skill directory; requesting files outside that scope should be rejected.","For deployments using HTTP transport, the server is likely bound to your configured environment/host; ensure you don't expose it unintentionally on public networks."]}}