MCP server providing Code Mode execution tools for API discovery and execution, with support for REST APIs, GraphQL, and SPARQL endpoints. Includes catalog search, dynamic code execution in sandboxed V8 isolates, auto-staging of large responses, and structured citation/provenance tracking.
Server defines 5 tools with reasonable descriptions and schemas, but exhibits significant gaps in parameter documentation, error handling guidance, and output schema clarity. Tool names follow verb_noun convention (search, execute) but descriptions are verbose and lack actionable recovery guidance. Schemas present but incomplete, many parameters lack type constraints and validation rules. No evidence of error classification, confirmation patterns for destructive operations, or per-item success/failure reporting. Output schemas are not documented.
Execute JavaScript code against an API in a sandboxed V8 isolate. Code runs with: searchSpec(query) / searchPaths(query) — search the API catalog or OpenAPI spec; listCategories() / listTags() — inspect categories/tags; getEndpoint(path) / getOperation(idOrPath) — get endpoint docs; api.get(path, params) for GET, api.post(path, body, params) for POST; Large responses (>100KB) are auto-staged into SQLite; Use limit/pagination params to control response size. Path params like /lookup/{id} are auto-interpolated from params.
Execute JavaScript code against a GraphQL endpoint in a sandboxed V8 isolate. Code runs with: gql.query(queryString, variables?) — execute GraphQL queries; schema.types(), schema.type(), schema.search() — introspection helpers; db.stage(), db.queryStaged() — staging helpers; Optional REST capability: api.get(path, params) / api.post(path, body, params) when restApiFetch is configured (hybrid GraphQL+REST mode); console.log() capture; Large responses (>30KB) are auto-staged into SQLite.
Execute JavaScript code against a SPARQL endpoint in a sandboxed V8 isolate. Code runs with: sparql.query(queryString, opts?) — run SELECT/CONSTRUCT/DESCRIBE; returns array of plain { variable: value } rows; sparql.ask(queryString) — run ASK; returns boolean; sparql.raw(queryString, opts?) — return unparsed { head, results } envelope; prefixes — object of common ontology prefixes (rdfs, owl, xsd, skos, dcterms, void, obo, uberon, go, ncbigene, efo, obi, sio, up, ensembl, bgee); prefixHeader — pre-built PREFIX block to prepend to queries; console logging capture; Large SPARQL responses (>30KB) are auto-staged into SQLite.
Search the API catalog (catalog mode) or OpenAPI spec (OpenAPI mode). Returns matching endpoints with full parameter documentation. Use this to discover API capabilities before calling {prefix}_execute. In catalog mode: accepts query/category/max_results parameters and performs keyword-based search over static ApiCatalog. In OpenAPI mode: accepts code parameter for agent-written JavaScript with full resolved OpenAPI spec and helper functions available.
Output schemas not documented. Tools return complex nested structures (staged datasets, introspection results, query responses) but no schema definition visible for LLM planning. LLMs cannot predict what fields to extract or how to chain results to downstream tools.
Error handling lacks recovery guidance. No evidence of error classification (retryable vs user-fixable vs fatal), actionable error messages, or suggestions for next steps. E.g., if a GraphQL query fails, the LLM has no guidance on whether to retry, adjust the query, or call a different tool.
Parameter 'code' in execute tools accepts arbitrary JavaScript but lacks validation rules, timeout constraints, or sandbox escape prevention documentation. No mention of what happens if code runs >30s, allocates >1GB memory, or attempts network access outside the API.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 41 | 2026-07-28+ | v2 |
Search the GraphQL schema or discover GraphQL capabilities. Returns matching types, fields, or operations. Use this to explore the GraphQL API before calling {prefix}_execute.
Descriptions are verbose (300+ chars) and bury key details. E.g., catalog-search description spans 4 paragraphs; LLMs may miss the core action. Descriptions should be 50-200 chars with examples in parameter docs, not tool description.
No confirmation or dry-run pattern for execute tools. Code execution is a state-modifying operation (can POST, DELETE, or trigger side effects via api.post). Agents should be able to preview results or confirm before committing.