adversarial_review

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

With no annotations, the description carries the full burden; it usefully discloses that the reviewer is a different model family with an adversarial posture, which is real behavioral context. It omits cost/latency implications of an external model call, any prerequisite such as resolving the model ID via list_models, and what the response looks like.

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?

Two sentences, no filler, with the unique value (different model family) and the recommended usage front-loaded. Every clause earns its place.

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 6-parameter tool with zero required fields and no output schema, the description leaves the agent guessing about input mode selection (inline content vs files) and the shape of the returned review. The core value proposition is well covered, but operational completeness is only adequate.

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 100%, so the baseline is 3, and the description adds essentially no parameter-level detail. It never clarifies that 'files' and 'content' are alternative inputs or that 'focus' narrows the critique, even though those are the choices an agent must make.

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+resource ('adversarial second opinion from a different model family (Gemini Pro)') and characterizes the output as flaw/edge-case/security hunting, which clearly separates it from siblings like analyze_files or deep_search. An agent can tell what this tool uniquely produces.

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 prescribes contexts ('ALWAYS use this for plan critiques, design reviews, and pre-merge code review'), which is strong positive routing. It stops short of naming alternatives — e.g. when plain analyze_files suffices instead — so the agent has no explicit when-not guidance.

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