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

Annotations only flag this as a non-read-only, non-destructive write. The description goes well beyond that: no training or spending occurs here, spend requires an out-of-band approval gate, the actual training happens in an external runner, and there is a strict one-candidate-plus-one-rerun cap per project. This is substantial behavioral context an agent could not infer from the annotations.

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-loads the core purpose, then layers prerequisites and limits with em-dash asides. Dense but every clause earns its place; slightly stream-of-consciousness phrasing keeps it from a 5.

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 with no output schema, the description covers scope ('Scope: write'), prerequisites, the external training mechanism, and rerun conditions. 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 100%, so all six parameters (including the superadmin-only customerId and the conditional rerunReason) are already documented, and the description's mention of base model, method, spend cap and rerunReason largely restates them. Baseline 3 is appropriate since the schema carries the parameter-level detail.

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 ('record the intent to train') and resource ('one candidate on the project's frozen dataset version'), then immediately disambiguates from what it is not: 'Trains nothing and spends nothing.' An agent can distinguish it from get_training_job and trigger_optimization without opening a 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?

Explicitly states prerequisites (a training_spend approval obtained via request_approval, decided by a human or approve-scoped key), the actor that actually trains (the runner outside the app), and per-project cardinality limits (one candidate plus one diagnosed rerun, which requires rerunReason). When-to-use and constraints are fully spelled out.

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