kb_search

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Describes freshness stamp ([fresh], [evidence changed]) and that absence notes are re-checked live. Results are inlined in full. No annotations provided, so description carries full burden; sufficiently transparent for this read-only search tool.

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?

Well-structured: first sentence states purpose, then usage guidance, then result details. Every sentence adds value. Slightly verbose but remains clear and front-loaded.

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?

Given no annotations or output schema, the description covers purpose, usage, result format (freshness stamp), and fallback behavior. It also guides on re-verification. Complete for a search tool with one parameter.

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?

Only one parameter 'query' with schema coverage 100%. Description adds value by providing examples ('symptoms work: "logout loop after refresh"') and mentioning the 'over-supply rule' from find, which guides the agent to provide all salient tokens.

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?

Clearly states it searches the repo's NOTEBOOK for durable notes from past agents, specifies search terms (plain task words, symptoms, symbol names, file names), and distinguishes from sibling 'find' by recommending use before find.

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?

Explicitly instructs to try this BEFORE find when the task might have been seen before, explains that hits can answer or point to files, and describes fallback behavior ('No hits or an empty notebook → fall through to find, no tax'). Also mentions re-verification with kb_write if evidence changed.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.