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

Beyond the annotations, the description discloses important side-effect boundaries: it never sends events, never decrypts the upload, never resumes a job, never grants a retry, and revoked/expired data stays erased. That erasure/retention disclosure is exactly the kind of context annotations cannot carry. The gap is that destructiveHint=true is never explained — the description reads as pure preservation, yet nothing says what is destroyed or overwritten when this save occurs.

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?

Three short sentences, no filler, with the core action front-loaded and the exclusions grouped afterwards. Sizing is appropriate for the tool; the only weak spot is that the middle sentence packs four distinct negations into a single run-on list.

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?

An output schema exists, so return values need not be explained, and annotations cover the safety profile in part. But for a destructive, idempotent, open-world write with six parameters, the definition never explains what is destroyed, how request_key governs replay, or what operation_id refers to — leaving real gaps an 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 description coverage is only 50%, and three parameters (work_id, request_key, arguments.operation_id) have no schema documentation whatsoever. The description adds no parameter meaning at all — it never mentions work_id, the operation_id that identifies the interrupted apply, or the idempotency role of request_key, so it fails to compensate for the coverage gap.

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 ('Save the original captured conversion results') plus a triggering condition ('after interrupted apply'), which is far more precise than the title alone. It is still somewhat opaque about what 'captured conversion results' are and how they differ from the many conversion-related siblings (conversion_result, conversion_review, conversion_batches), but an agent can tell this is a persistence step rather than an inspection step.

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 phrase 'after interrupted apply' gives a usable trigger, and the negation sentence rules out several incorrect uses (it does not resume a stopped job or grant a retry), which is genuine when-not guidance. However, no alternative tool is named, and 'interrupted apply' is never defined, so an agent must infer which sibling to call for the resume/retry case it is told to avoid.

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