recheck_apply

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

Annotations only give generic hints (mutation, non-idempotent, closed-world); the description adds the actual side effects: retrieved_at is set to now, git preserves the prior claims, `by` appends a §5.2 `verified` entry, and stale_days>0 advances stale_after. That is real disclosure beyond annotations and consistent with idempotentHint=false. It does not say what happens on partial/mismatched results or errors.

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?

Every clause carries information, but the text is semicolon-fragmented and jargon-dense ('§5.2', '--attest', 'git keeps what they said before'), making it harder to parse than the content requires. The purpose is front-loaded, but the ordering of side-effect clauses is not.

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 6-parameter mutation tool with no output schema and no schema descriptions, the definition explains the workflow prerequisite, the result shape, and each mutation's effect. Absent are return/error behavior and whether the write is scoped to a workspace or bundle by default, but these are modest for a tool whose annotations already flag it as non-destructive and non-idempotent.

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 0%, so the description must carry the load and largely does: results is described as a list of {bundle?, file, id?}, accepts a bare file=/id= form, omitting id means every source on the concept, `by` triggers the verified entry, and stale_days>0 advances stale_after. Only `bundle` scoping and the interaction between results and file/id are left thin.

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?

Specific verb+resource: 'Record a re-check' after comparing due sources with the claims citing them. It names the sibling that must run first (recheck_plan) and describes the artifact written. It stops short of distinguishing itself from adjacent verbs like 'verify' or 'record' in the sibling list, and key terms ('due source', 'concept', 'bundle') are left undefined.

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?

Explicit precondition — use only after you have compared each due source against its citing claims via recheck_plan and corrected what changed. That sequencing is exactly the guidance an agent needs. No exclusion is stated for verify/record/status, which is the only real gap.

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