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

With no annotations, the description carries the full behavioral burden and does much of it: seven-day default expiry, edit semantics for omitted expiry, the requirement for fresh operator authorization on revoke-and-replace, the API-key requirement, and the explicit statement that neither automation_mode authorizes charges. It does not describe what happens to an existing budget if a second one is created for the same parent, or whether the call is idempotent – meaningful gaps for a mutation tool.

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 purpose and the authorization precondition are front-loaded, then editing rules, then the negative scope. Six tight sentences with no filler. It is dense but every sentence carries a distinct constraint.

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 10-parameter mutation tool with no annotations and no output schema, the description supplies an unusually strong safety and lifecycle picture (authorization, expiry defaults, non-effect on escrow). It falls short only on the undocumented optional parameters and on what the call returns, though the absence of an output schema makes the latter less critical.

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?

At 40% schema coverage and 10 parameters, the description covers some of the gap (expires_at defaulting, grant_id for edits, per-task vs. total limits), but leaves enabled, allowed_categories, max_delegation_depth, and parent_task_id unexplained in both places. The tool's most safety-relevant numeric fields are only partly illuminated, so it neither keeps the baseline nor fully compensates.

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 gives a specific verb and resource with scope: 'Set your operator-approved total and per-task limits for separately funded subtasks of a funded parent job.' It distinguishes the create vs. edit path ('Pass grant_id when editing') and implicitly separates itself from revoke_delegation_budget by noting that revoking requires fresh authorization. It stops short of naming the sibling tool explicitly, so an agent must infer the pairing.

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?

There is explicit gating ('Never use without your operator's authorization') and clear branching for edit vs. create ('Pass grant_id when editing; omitted expiry preserves the current expiry'). Exclusions are stated ('Does not move parent escrow or authorize automatic charges'), which rules out misuse. What is missing is a direct pointer to revoke_delegation_budget or other siblings for the revoke-then-recreate workflow.

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