Loading ToolBench
Two tools with clear action verbs and reasonable descriptions (194 - 250 chars, within baseline 34 - 392). Input schemas present with types and defaults. However, output schemas are completely undocumented, LLMs cannot plan downstream calls or extract structured data. Parameter descriptions lack format constraints (e.g., 'max 400 chars, 50 words' stated but not enforced via schema). No error handling guidance. Rate limiting and API key validation are server-side, but no tool-level error recovery hints. Composition is sound (web vs. local search separation), but missing pagination metadata in output and no guidance on when to retry vs. fail.
Searches for local businesses and places using Brave's Local Search API. Best for queries related to physical locations, businesses, restaurants, services, etc. Returns detailed information including: - Business names and addresses - Ratings and review counts - Phone numbers and opening hours Use this when the query implies 'near me' or mentions specific locations. Automatically falls back to web search if no local results are found.
Performs a web search using the Brave Search API, ideal for general queries, news, articles, and online content. Use this for broad information gathering, recent events, or when you need diverse web sources. Supports pagination, content filtering, and freshness controls. Maximum 20 results per request, with offset for pagination.
Output schemas completely undocumented. Tools return formatted text strings, but LLMs cannot parse structured fields (title, url, rating, phone, hours) for downstream chaining. Breaks pattern:tool-chain and pattern:response-shaper.
Parameter constraints stated in descriptions ('max 400 chars, 50 words', 'max 9 offset') but not enforced in JSON Schema. LLMs cannot read prose constraints reliably, use minLength, maxLength, minimum, maximum, enum, or pattern fields.
No error recovery guidance. Rate limit errors, API failures, and fallback behavior (local→web) are handled server-side but not communicated to LLM. Missing pattern:recovery-guide and pattern:error-classification.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 53 | 2026-07-28+ | v2 |
Pagination support (offset, count) present in schema but no documentation of total results, next_cursor, or result limits. LLMs cannot determine if more results exist or how to iterate safely. Violates pattern:paginated-result.
brave_local_search description mentions fallback to web search but does not document what the LLM should expect (same format? different fields?). Undocumented behavior breaks agent reasoning.