product_code_statsFDA review-time statistics (median and 90th percentile days to decision), clearance count and recall count for a product code.
| Name | Required | Description | Default |
|---|---|---|---|
| product_code | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already convey read-only and non-destructive behavior. The description adds useful return-value context (specific metrics), but does not disclose data scope, time period, formatting, or any rate/access limitations. It does not contradict the annotations.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
A single sentence with no filler. It front-loads the tool's purpose and immediately states the key output metrics. Every word contributes value.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a simple one-parameter read-only tool, the description identifies the required input and the key return values. It lacks deeper context such as data source or valid code enumeration, but overall it gives enough for correct invocation.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
With schema description coverage at 0%, the description must define the parameter. It only restates 'product code' without explaining what constitutes a valid code, its format, or how to discover it. The description offers little beyond the parameter name itself.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the resource (product code) and the exact statistics returned (median and 90th percentile days to decision, clearance count, recall count). It lacks an explicit verb but is unambiguous, though it does not differentiate from siblings like get_clearance or search_recalls.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description gives no explicit guidance on when to use this tool versus alternatives. Siblings such as get_clearance and search_recalls likely overlap in clearance and recall data, but the description does not mention exclusions or selection criteria.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Add one secure layer between your agents and this server.
Each tool has a clear primary purpose: AI device list, predicate candidates, monitoring feed, clearance detail, product code stats, recall search, and general full-text search. Minor overlap exists between find_predicates and product_code_stats (both product-code driven) and between search_recalls and search_records, but the descriptions distinguish specialized from general use.
Most tools follow a verb_noun pattern like find_predicates, get_changes, get_clearance, search_recalls, and search_records. Two names — ai_devices and product_code_stats — are noun phrases, creating a minor inconsistency, but the overall pattern remains readable and predictable.
Seven tools is well-scoped for a read-only FDA data server, covering lookup, search, statistics, monitoring, and specialized research use cases without unnecessary duplication. Each tool earns its place.
The set covers clearance details, recall search, product code statistics, change monitoring, predicate research, and the AI device list. There is no dedicated clinical trial tool, though search_records includes trials, and no device classification lookup, but these are workable gaps rather than dead ends.