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

Annotations only declare readOnlyHint=false, destructiveHint=false, openWorldHint=false, so the description's job is to say what the write actually causes — and it does: "A built idea is never offered as a feature again: when the site says it too it is left out, otherwise it becomes saying it on the site." That is genuine downstream-effect disclosure an agent could not get from the annotations. It omits idempotency (what re-checking the same key does) and any return/confirmation signal, keeping it below 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?

The text is not front-loaded: it opens with a caveat about what the ideas study can and cannot see rather than the action the agent must take. Sentences are long and run together, the parameter name is wrong, and the trailing "Free." is a dangling fragment that adds nothing. The information is mostly useful but the packaging is muddled.

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 write tool taking a nested array of up to 20 check objects with no output schema, the description covers the intent and the consequence chain but not what happens on an unknown/invalid key, whether a resubmitted key overwrites the prior verdict, or what the caller gets back. Adequate to attempt a call, insufficient to call it confidently in edge cases.

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?

The description supplies the conditional logic the schema does not spell out: `file` is only meaningful for a built idea, `said` is only meaningful for a built idea, and `built: false` means the product does not do it. It does, however, refer to "ideasToCheck," which is not a parameter name (the actual field is `checks`), creating a mapping ambiguity, and it never mentions the 1–20 item bound.

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 never states plainly that this tool records/uploads the repo-check verdicts for a batch of ideas; the operative instruction is buried in "pass what you found" inside a sentence about repo searching. The rationale sentence ("The ideas study reads only the site, so it cannot know what the product already does") explains why the tool exists but not what it does. An agent can infer the purpose from the name plus the field semantics, but only after work.

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 an explicit precondition and ordering: search the founder's repo for each idea workspace in ideasToCheck (or any idea before offering it), then pass the findings here. That is a real when-to-use trigger rather than an implication. It does not name sibling alternatives (ideas_run, ideas_skip, repo_confirm) or state what to do if the repo cannot be searched, so it stops short of a 5.

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