create_pull_request_review_comment

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

The annotations are sparse (readOnlyHint=false, idempotentHint=false, destructiveHint=false) which is ambiguous. The description adds two meaningful behavioral facts beyond the annotations: the comment is created on the exact head SHA (concurrency constraint), and 'the body is sent through stdin and is not audited.' The stdin/audit disclosure is genuinely valuable behavioral transparency. However, it doesn't disclose what part of the PR object gets modified, whether existing review threads are reused, or other side effects.

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, both dense with information. The first sentence conveys purpose and scope, the second adds a security/audit transparency note. No filler words. Could arguably add one more sentence on parameter relationships, but for what it covers it's efficient and 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?

For a 12-parameter, 6-required tool with no output schema and 8% schema coverage, the description is notably incomplete. It does not explain how line/startLine/side/startSide interact to form ranges, the meaning of expectedHeadSha as a safety mechanism, subjectType implications, or what the response returns. However, the audit-notice is a nice touch. Given the tool's moderate complexity (many siblings in the same domain), the description falls short of what an agent needs to correctly 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?

Schema description coverage is only 8% (only 'repository' has a description in the schema). The tool has 12 parameters with complex interrelationships (e.g., line/startLine/side/startSide for ranges, subjectType for line vs file, expectedHeadSha for concurrency). The description does not explain any parameter semantics beyond what the schema enum values already imply. For a high-parameter tool like this, the description should compensate, but it doesn't uncover the meaning of required parameters like expectedHeadSha or the line/startLine relationship.

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 ('Create an inline line- or file-level review comment') with a specific resource (pull request, with head SHA constraint). It distinguishes line-level vs file-level via subjectType. However, it doesn't explicitly differentiate from sibling tools like review_pull_request or comment_pull_request which could have overlapping purposes, so it's clear but not fully differentiating.

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 notes the comment is created 'on the exact expected pull request head SHA', which provides some context about the constraint. However, it doesn't explain when to use this vs reply_pull_request_review_comment, review_pull_request, or comment_pull_request. No exclusions or alternatives are named. The head SHA constraint hints at a concurrency-safety use case but isn't explicitly framed as guidance.

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