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

Annotations already flag destructiveHint=true and readOnlyHint=false; the description adds the safety gating (remove deletes only with confirm true after confirmation, add only stores approved facts) and discloses the four possible return states with their follow-up actions. It does not state idempotency or rate behavior, and it defers important detail to 'the manual'.

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 and the five step kinds, and every clause carries operational weight. The sentences are long and semicolon-chained, which slightly hurts scannability, but there is no filler.

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?

No output schema exists, so the description must carry return semantics — and it does, naming all four states (waiting/running/done/failed) and the correct next action for each. Combined with the deletion and confirmation rules, an agent has what it needs to call and follow up correctly, with only minor gaps around error and limit behavior.

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?

With 67% schema coverage the schema already documents most body fields, but the description supplies the cross-parameter workflow the schema cannot express: proposal_id and confirmation come from propose's result, confirm is only valid post-confirmation, and parts must be sent in order. It adds little on `store` beyond what the schema already says about read-only public links.

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?

It names a concrete verb (start) and resource (a step in a private store) and enumerates the five step kinds it drives: propose, add, documents, ask, remove. It falls short of full sibling differentiation — notably it never disambiguates the `ask` step from the sibling tool named `ask`, which an agent could easily confuse.

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 gives per-step conditions: propose stores nothing, add requires propose's proposal_id plus the person's quoted reply, remove requires confirm true only after the person confirmed. It also routes follow-up calls ('call getStep' when running), though it never states when to reach for this tool versus getStep/getManual/createStore as a starting point.

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