Search Code

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

Annotations already declare readOnlyHint=true and openWorldHint=false, and the description adds real behavioral context beyond that: results are meaning-ranked chunks with confidence scores, coverage spans all indexed projects, and cost is claimed 50-90% lower than reading full files. The one blemish is the coercive 'REQUIRED — call this FIRST for ANY code question' framing, which risks over-invocation and adds no descriptive information.

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 front-loaded: the imperative, then scope, then triggers, then rationale, then workflow, then the naming caveat — an agent reading only the first line still routes correctly. Slightly verbose: the trigger list is long and its final clause ('or ANY reference to project code') subsumes everything enumerated before it, and the WHY/token-savings section reads as advocacy rather than specification.

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 four-parameter read-only search with no output schema, the description covers return format, ranking semantics, scope, and the downstream chain, so an agent can call it correctly. Remaining gaps are minor: no guidance on what a confidence score threshold looks like in practice or how 'group' interacts with 'project' filtering.

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 coverage is 100%, so the baseline is 3, but the description earns an upgrade by clarifying project naming: directory basenames, not org/repo form, with a concrete example ('billing' not 'org/billing') — a convention the schema's 'Project name or "all"' does not convey. The group parameter and limit bounds are left entirely to the schema.

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 and resource — semantic code search across all indexed projects — and immediately defines the return shape (ranked code chunks with symbol definitions and usages). The workflow sentence positions it as the entry point ahead of get_chunk and find_usages, so an agent can tell where it sits relative to siblings 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?

Explicitly says when to call it ('call this FIRST for ANY code question') and enumerates trigger intents: location, implementation, how-X-works, counts, versions, patterns, flows, error handling, auth, API usage. It also hands off explicitly: search → evaluate scores → get_chunk for details → find_usages for dependencies, which names the alternatives and the condition selecting each.

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