Search workspace

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

Annotations already declare readOnlyHint=true and openWorldHint=false, so the safety profile is covered; the description goes further with scan cost ("medium (file scan)"), a 3s TTL cache scoped to identical queries, batch-safety, and the output fields (hits, complete, diagnostics, hint) despite no output schema. It stops short of explaining truncation/limit behavior or whether hits are truncated when `complete` is false, which would matter for correctness.

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?

Purpose first, then labeled Use when / Avoid / Speed / Output / Batch-safe blocks with zero filler. Every line carries distinct information, and the most decision-relevant content is 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?

Good on outputs (pre-empting the missing output schema) and cost, but for a 12-parameter search tool the absence of any guidance on scoping (glob/relative_path), result limiting (limit, count_only), and context_lines leaves the agent likely to call it with defaults that don't match its intent.

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?

Schema description coverage is 8% (only relative_path is documented), so the description carries the burden for 12 parameters but only adds meaning to `query`, `kind`, and `regex`. Filters like glob, exclude_glob, whole_word, case_sensitive, include_hidden, context_lines, count_only, and limit are named nowhere, leaving the agent unaware they exist.

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?

States a specific verb plus the four exact search modes (file name, directory name, content literal/regex, symbol definition lines), which maps directly onto the kind enum. The agent can tell this apart from project_read_file and project_list_files without opening any schema.

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?

"Use when: ... use kind=literal_content or definition" and "Avoid when: you already know the exact path to read" give both the trigger and the exclusion, and the exclusion implicitly routes to project_read_file. No inference is required.

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