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

Annotations already declare readOnly, idempotent, non-destructive and closed-world, so the safety profile is covered. The description adds genuine context beyond that: the tool emits a line for decisions.jsonl and, when a durable rule is stated, also produces a learning candidate with the location to persist it. Note a mild tension: 'Record' reads like a write, but 'Returns a line for decisions.jsonl' reconciles it as a read-only composition step.

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?

Two short sentences that front-load the action and the required reasoning, then disclose the return shape. Dense but every clause carries information; there is no filler or hedging.

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?

With an output schema and full parameter coverage, the description needn't explain return fields, and it correctly focuses on the two behavioral facts an agent could miss: the reason requirement for skipping the top option, and the dual output (journal line plus optional learning candidate). The only gap is no guidance tying it to the neighbouring outcome tools.

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 the baseline is 3. The description restates what the schema already documents (considered is best-ranked first; skipping the top option needs a rationale) rather than adding format, ordering or validation insight the agent could not read from the schema.

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 pairs a concrete verb with its resource ('Record what the owner chose') and enumerates exactly what gets captured: choice, options in rank order, and what was passed over plus why. It is specific enough to separate from record_outcome_experiment, but it never names that sibling or any other, so the differentiation is inferential rather than explicit.

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 one real conditional rule ('Passing over the top option needs a reason'), which tells the agent a precondition for a valid call. However, nothing states when to use this versus record_outcome_experiment or advance_outcome, and the critical gate 'only after the owner actually said so' lives only in the schema, not the description.

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