Loading ToolBench
FDA Regulations MCP — US Food & Drug Administration regulations (21 CFR). A thin, FDA-branded, keyless wrapper over the official eCFR API, scoped to the whole of Title 21.
Two tools with strong, detailed descriptions (194 - 250 chars each) that clearly explain use cases and scope. Input schemas are present with proper JSON Schema structure and parameter descriptions. Tool names follow verb_noun convention (fda_regulation, fda_search). However, output schemas are not documented, LLMs cannot predict response structure for downstream planning. Parameter descriptions are good but lack explicit constraints (e.g., limit range 1 - 20 is mentioned in description but not in schema). No error handling guidance or recovery paths documented. No tool annotations (readOnlyHint, etc.) despite both tools being read-only.
Get the full text of one FDA regulation — a US Food & Drug Administration rule codified in 21 CFR — by its citation. Returns the exact regulatory wording currently in force. Answers "what does 21 CFR 211 require", "what is the FDA regulation for X", "does the FDA require X", "read 21 CFR 211.100", "the FDA good manufacturing practice rule". Forgiving citation input: "211.100", "21 CFR 211.100", "§211.100", even "211.100(a)" (paragraph stripped to the section). Covers 21 CFR part 211 current good manufacturing practice (cGMP / GMP) for finished drugs, part 210 cGMP general, part 820 quality system regulation / QMSR for medical devices, part 801 medical device labeling, part 101 food labeling / nutrition facts, part 314 new drug applications (NDA/ANDA), part 807 device establishment registration, part 111 dietary supplement cGMP, part 1308 controlled substance schedules — the whole of Title 21 (food, drugs, devices, cosmetics, biologics). This is FDA REGULATIONS (the rules/regulatory text); for FDA DATA (drug labels, adverse events, recalls) use the openfda tools. Pass a whole part (e.g. "211" or "820") to get that part's section list. Example: fda_regulation({ citation: "211.100" }) -> written procedures / process control; fda_regulation({ citation: "21 CFR 101.9" }) -> nutrition labeling. Keyless.
Keyword search across FDA regulations — US Food & Drug Administration rules in 21 CFR. Answers "what FDA regulations cover X", "the FDA regulation / rule about X", "find the FDA requirement for X". Great for topics: good manufacturing practice (GMP / cGMP), quality system regulation, medical device labeling, drug labeling, nutrition facts / food labeling, new drug applications, dietary supplements, cosmetics, biologics, controlled substances, current good manufacturing practice for drugs and devices. Returns matching FDA regulations with citation (21 CFR), heading, excerpt, and source URL. This searches FDA REGULATIONS (regulatory text); for FDA DATA (drug labels, adverse events, recalls) use the openfda tools. Example: fda_search({ query: "medical device labeling" }); fda_search({ query: "good manufacturing practice", limit: 15 }). Keyless.
Output schemas not documented. LLMs cannot predict response structure (fields, types, pagination) for planning downstream calls or extracting data.
No tool annotations (readOnlyHint, destructiveHint, idempotentHint). Both tools are read-only and idempotent, annotating them enables agent safety checks and retry logic.
No error handling guidance. Errors (e.g., 'citation not found', 'eCFR timeout') lack recovery hints. LLMs cannot self-correct or know whether to retry.
Parameter constraints not formalized in schema. 'limit' description says '1-20' but schema lacks minItems/maxItems. LLMs may pass invalid values.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | C | 62 | 2026-07-28+ | v2 |
Transport mechanism unknown. Source code shows custom fetch logic but no explicit MCP transport binding (HTTP, SSE, STDIO). Cannot verify protocol compliance.