delegate_patch

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

With no annotations provided, the description carries the full burden of disclosure. It meaningfully warns that the tool only proposes a patch and does not apply it, and it outlines Claude's responsibility to inspect, validate, apply, format, test, and review. This is a critical behavioral trait that goes beyond the tool's name. It doesn't cover costs or rate limits, but the main surprise factor is addressed.

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?

The description is two sentences long. The first sentence conveys the core purpose, and the second delivers the essential caveat about not applying changes. Every word is meaningful; there is no filler or redundancy.

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 tool is well-scoped with an output schema, and the description clearly explains what the tool does and what it does not do. The missing guidance on provider selection and optional parameters is a minor gap, but the core context needed for a delegation tool is present. Overall, it is fairly complete for an agent to assess whether to invoke it.

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?

The schema description coverage is only 40% (task and context have descriptions), yet the tool description adds no parameter-level guidance. It doesn't explain what 'provider' options mean, what 'reasoning_effort' controls, or how 'max_output_tokens' should be set. With low schema coverage, the description was expected to compensate but did not.

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 states a specific action ('Ask an external model') and a specific deliverable ('a minimal unified diff'), and explicitly notes that the tool does not apply changes. This clearly distinguishes it from the sibling tools delegate_analysis and delegate_review, which likely focus on analysis and review rather than patch generation.

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 gives clear context on when to use this tool: when an external perspective on a minimal diff is needed. It also states a key exclusion ('does not apply changes') and the follow-up steps Claude must take. However, it does not explicitly name alternatives like 'delegate_analysis' for analytic tasks, so it lacks explicit when-to-use-other-tools guidance.

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