review_reject

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

Annotations declare readOnlyHint=false, destructiveHint=true and openWorldHint=false, so the write/non-open nature is already known. The description adds real value beyond that: the item is retained, nothing is deleted, and a user-supplied reason is stored. It does not resolve the mild tension with destructiveHint=true (i.e., whether a rejection is later reversible), which keeps this below a 5.

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?

Well front-loaded: purpose, then retention behavior, then the guardrail, then the write requirement. Slight redundancy in the repeated instruction about the user's words and the clause about not approving one's own suggestions, which belongs more naturally to the approve sibling.

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 5-parameter mutation with no output schema, the description covers the important gaps: what happens to the item, the authorization precondition, and the project requirement. It could say what happens on success or whether a rejection can be undone, but nothing essential to correct invocation is missing.

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 80%, so item_id, reason, model and project are largely documented already (baseline 3). The description reinforces two constraints that matter operationally: user_request must actually quote the user's words, and project="<name>" is mandatory for every write, which is stronger than the schema's generic "Required for every write" note.

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?

States a specific verb and resource ("Reject a review item") and immediately defines the resulting semantics: the item is kept and marked rejected with the user's reason. Together with the sibling review_approve, an agent can tell exactly which of the two decision tools this is without opening the schema.

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?

Explicitly states the trigger condition and the prohibition: call ONLY when the user has explicitly asked in this chat for a decision on this specific item, and never act on the agent's own initiative. The approve/reject decision context routes the agent appropriately and leaves nothing to inference.

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