Approve a pull request to make it mergeable, or reject it with a logged reason for decision training, while blocking self-reviews.
审阅 PR——approve 进 reviewed（可合并）；reject 终态 rejected（拒因进 decision-log 负样本训练信号）。提交者不可自审（利益冲突拒绝）；verdict 精确匹配 approve/reject（大小写敏感）。
| Name | Required | Description | Default |
|---|---|---|---|
| note | No | 审阅意见（reject 时即拒因） | |
| pr_id | Yes | PR 标识 | |
| verdict | Yes | 审阅结论（精确匹配，区分大小写） | |
| data_dir | No | 数据根目录 | |
| reviewer | Yes | 审阅者（不可为 submitter） |
Changes observed during successful MCP inspections.
v0.1.0Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations present, the description carries full responsibility and delivers: approve/reject state changes, terminality of reject, decision-log negative training signal, conflict-of-interest rejection, and exact case-sensitive verdict matching. This is substantial behavioral disclosure for a mutation-style review action.
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?
A single dense sentence front-loads the action and then packs consequences, constraints, and exact matching into semicolon-separated clauses. No filler or redundant restatement; every clause adds information.
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 simple 5-parameter action with no output schema, the description covers state outcomes, constraints, and parameter semantics adequately. It does not describe return values or error behavior, but these are less critical for invoking this review action correctly.
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%, giving the baseline of 3; the description adds workflow-level meaning by tying 'note' to rejection reason, clarifying 'verdict' must exactly match approve/reject case-sensitively, and reinforcing that 'reviewer' must differ from the submitter. It elevates parameter understanding beyond field names.
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 opens with a clear verb+resource ('审阅 PR') and specifies the two verdict outcomes (approve → reviewed/mergeable, reject → terminal rejected). This distinguishes pr_review from sibling tools like pr_submit and pr_merge by describing the review action's state transitions.
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?
It provides clear context for when to call this tool—when a PR needs review—and includes an explicit restriction (submitter cannot self-review). It does not explicitly name alternatives like pr_merge or pr_submit, so it falls just short of full alternative-routing guidance.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.