cos_doc_search

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

The description adds significant behavioral context beyond the annotations: it explains the stage-1 metadata pre-filter, how auto_context soft-defaults domain from the active task's swimlane, and that filter_hints are heuristic suggestions never auto-applied. It also discloses the two-stage search process and response meta fields. This goes far beyond the readOnly/idempotent annotations and provides rich behavioral transparency.

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?

Although lengthy, the description is well-structured and every sentence earns its place. It front-loads the core purpose, then systematically explains the pre-filter logic, parameters, and response meta. For a 9-parameter tool with no schema documentation, this level of detail is necessary and appropriately organized with an Args section and clear examples.

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?

The description is complete for the tool's complexity. It covers parameter semantics, behavioral edge cases (e.g., stale docs, superseded specs), response format (JSON envelope with listed fields), and the decision-making flow for filter hints. The presence of an output schema does not negate the need for this semantic richness, and the description delivers it fully.

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 0% schema description coverage, the description fully compensates by explaining each parameter (query, source_types, limit, mode, domain, layer, since_iso, include_inactive, auto_context) with examples, defaults, and behavioral nuances. For instance, it clarifies that 'domain' is 'Indexed' and 'include_inactive' hides chunks marked is_active=0. This is a model of parameter documentation.

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 tool performs 'Semantic + lexical search over project documentation chunks.' It specifies the resource (project documentation) and distinguishes itself from generic search tools like cos_search by emphasizing metadata pre-filtering and stage-1 filtering. The title 'Search Project Documentation' reinforces the purpose.

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 provides clear context for when to use the tool, including a mental model ('Filter → Search → Summarize') and specific guidance for parameters like since_iso ('Use when the agent asks about recent or current state') and include_inactive ('Set True for decision-history retrieval'). It does not explicitly name alternative tools or exclusions, but the usage context is well implied for a documentation search tool.

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