Read GitHub pull requests

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

Annotations only say readOnly/idempotent/openWorld; the description adds substantial behavior: a cancelled check counts as failed, checks pooled for up to wait seconds with waited.ended_by "timeout" and a hint to call again, patches capped at 60000 characters with patch null and patches_truncated afterwards, at most 3 failed-job logs, and truncation signalling on list. It also flags that PR text is untrusted data rather than instructions, which is a real safety behavior not encoded anywhere in annotations.

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?

It is dense and front-loaded by method, with every sentence carrying concrete facts rather than filler. The tradeoff is a run-on listing style with inline quoted enum values that takes effort to parse, and the truncation rule sits after the diff method where it is easy to miss.

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 17-parameter, four-mode tool this covers what an agent needs: which method to pick, how to format the PR reference, defaults and caps, polling/timeout semantics, and truncation behaviour. Return values are additionally backed by an output schema, so the description does not need to carry that burden.

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 adds meaning the schema does not: the accepted PR reference forms (12, #12, owner/repo#12, URL, explicitly not an issue's URL, bare number resolved against the working tree's origin), the refusal of repo:/org:/user: inside query, and cross-parameter behavior such as paths scoping limits together with patch truncation.

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 pairs a specific verb and resource ("Read GitHub pull requests") with a full enumeration of the four read methods and exactly what each returns (view: state, reviews, mergeability, etc.; checks: last commit checks; diff: changed files). An agent can distinguish this read-only tool from the mutation siblings (github_pr_merge, github_pr_edit, github_pr_review) purely from the text.

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 method-by-method breakdown gives clear intra-tool routing: use view for one PR in full, list for filtered enumeration, checks to poll CI, diff for patches. What it does not do is name cross-tool alternatives or exclusions (e.g., when to prefer github_pr_review or github_stack_read), so it stops short of 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.