start_session

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

Beyond the annotations, the description discloses the idempotency-key retry contract and that no model inference occurs, which is genuine behavioral context an agent needs. Worth noting a tension: 'replaying it returns the same session' reads as replay-safe while the annotations set idempotentHint=false, though the annotation likely refers to the create-new-resource default rather than request_id-keyed replay.

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?

Three short sentences, front-loaded with the core action, then the required-parameter constraint, then the capability boundary. Nothing is padded or redundant.

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 creation tool with no output schema, the description never says what the caller gets back — a session ID, an offer handle, expiry metadata, or a status object. Since no output schema exists to carry that, the return shape is an unresolved 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?

Schema description coverage is 100% and the request_id property already documents the caller-generated UUID, retry retention, and non-reuse rules. The description's mention of a 'unique request_id' and replay behavior only restates that, adding no new syntax or format detail beyond 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 opens with a specific verb+resource ('Create ... offer conversation') and adds a scoping qualifier ('isolated seven-day'), which lets an agent distinguish it from siblings like send_message or get_status. 'Offer conversation' is somewhat domain jargon, keeping it short of a 5.

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?

It supplies a retry rule ('replaying it returns the same session') and a negative capability note ('No model inference'), but never states when to call this versus its siblings or what prerequisite state must exist. Usage is implied rather than prescribed.

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