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

Annotations only supply readOnlyHint=false, so the description carries the real burden and does so: it discloses the atomic per-draft scope, the draft_changed failure mode where nothing is approved, the human-in-the-loop requirement that the person must have read the draft, the mandatory accusation_untrue attestation for Defamation, and the capability gate (cases.approve_own). That is far more than the annotation provides.

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?

Purpose and enabling condition are front-loaded, then failure handling, then the specialized Defamation rule in priority order. The dense parentheticals and long clauses make some sentences harder to parse than necessary, but every sentence conveys a distinct operational rule.

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 three-parameter mutation tool with no output schema, the description covers the precondition, the success condition, the failure token, and the conditional required nested parameter, which is nearly everything needed to call it correctly. The only omission is any statement about return values or post-approval state, which is acceptable absent an output schema but leaves a small gap.

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?

With 0% schema description coverage, the description must compensate and largely does: draft_id is defined as 'the draft you showed the person (get_case's draft.draft_id)' and the nested accusation_untrue.attested_by is explained as naming the person at the company who attests the accusation is untrue. case_id is left implicit, which is a minor but real gap given no field-level schema docs.

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 ('Approve filing one of your cases') plus the enabling condition ('when your company requires approval (set_approval_required)'), which cleanly separates it from sibling tools like mark_filed and set_approval_required. An agent can identify the operation 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?

Gives explicit when-to-use context (only when approval is required by the company setting), pins the approval to a single draft sourced from get_case, and prescribes the correct alternative path when draft_changed occurs ('show the person the newer draft and approve it only once they have read it'). It also carves out the Defamation precondition, so the agent knows a hard requirement before calling.

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