Self-check items

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

readOnlyHint=true already tells the agent this is a safe read, and the description adds genuinely useful context: the output is advisory only and excluded from the auto-approve gate, and the verdicts must be persisted in a specific `<!-- opendrs-self-check:v1 -->` block in the head commit message. It does not describe the shape of the returned collection itself, but behavior beyond annotations is substantially disclosed.

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 trigger is front-loaded, but the body is one sprawling semicolon-chained sentence with an embedded JSON example. Much of the length is downstream workflow (how to write verdicts) rather than describing the tool, which dilutes the definition even if some of it is useful context.

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?

With no output schema, the description does the work of naming what comes back (captain facts, reviewFirst hunks, check questions) and what each verdict means. It stops short of describing the concrete return structure (list vs. grouped object, id format), which leaves a small gap for a tool whose results must be parsed and mapped to verdicts.

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 base, head, repoDir and maxTokens are fully documented in the schema, including defaults and the 300-6000 budget range. The description adds nothing about any parameter, so baseline 3 applies.

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 resource is well identified – 'the ids (captain facts, reviewFirst hunks, check questions) from the review brief' – and ties it to the sibling review_brief conceptually. The verb is only implied ('call before opening...'), which is slightly indirect, but an agent can tell this is a retrieval of self-check items rather than a diff score or hotspot list.

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 trigger is stated outright: 'Call before opening or updating a pull request.' It also scopes the purpose (pre-PR self-check) and clarifies the tool is advisory and never part of the auto-approve gate, which prevents the agent from treating the verdicts as gating criteria. Nothing about when to invoke it is left to inference.

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