recheck_plan

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

Annotations already declare read-only, idempotent, closed-world and non-destructive; the description reinforces this with "Nothing is executed here" and adds real operational context the annotations lack — that the agent must supply its own tool permissions to replay calls, and that args come from the bundle. It also describes the result groupings (sources, cites, calls, manual). It leaves unclear what happens when a recorded call fails or is no longer permitted.

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?

Roughly five dense sentences, front-loaded with the selection criteria and then the action workflow. Every sentence carries information, though the 'one never retrieved is due once its concept is stale' parenthetical is convoluted enough to require a second read.

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 necessary work of explaining the return shape (sources/cites/calls/manual) and the follow-up path. It still leaves gaps for a two-parameter tool: the unaddressed `bundle` argument and undefined domain terms like `stale_after` that the agent must guess at.

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 carries the full burden. It does decode `older_than_days` as the N-day retrieval threshold and ties staleness to `stale_after`, but the `bundle` parameter is never mentioned and remains opaque despite a sibling tool named `bundles`.

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 identifies a specific resource and scope: facts due for re-check, selected either by retrieval age (N days) or by concept staleness. It also implicitly distinguishes itself from recheck_apply, the tool that records the check. It never states a leading verb ("list"/"plan"), so the read-only planning nature must be inferred from the closing sentence.

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 gives concrete workflow direction: replay each `calls` entry yourself, read the recorded args first, open `manual` resources, correct stale claims, then record with recheck_apply. The handoff to recheck_apply is explicit. It does not say when recheck_plan should be skipped or how it relates to siblings like search/verify/lint.

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