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

With annotations only covering safety/idempotency hints, the description discloses substantial extra behavior: decisions are final and non-rewritable (consistent with idempotentHint=false), the approval executes nothing itself, and the decision is stamped with the calling key's id and scope. This is exactly the audit and side-effect context an agent cannot infer from the schema.

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?

Front-loaded with the core action, then finality, side effects, and scope in tight, information-dense sentences. Every sentence carries distinct meaning with no filler or repetition of the schema.

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 mutation tool with no output schema, the description covers the essential unknowns: finality/irreversibility, that it triggers no execution itself, and the exact permission scope required. Nothing an agent needs to call it correctly is missing.

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 already 80%, covering note, projectId, approvalId, and customerId; the description only loosely maps 'training spend or the production switch' onto the state enum. It adds little parameter meaning beyond what the schema provides, so the baseline of 3 applies.

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 pair (approve/decline) and a precise resource (a requested fine-tuning approval, either training spend or the production switch). An agent can distinguish this decision action from the surrounding request/list/record siblings 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 a concrete precondition ('an already-decided approval is refused') and a hard scope gate ('write, admin and superadmin keys cannot call it'), so the agent knows when it is eligible to call. It does not name an explicit alternative (e.g., request_approval) for creating rather than deciding, which keeps it 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.