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

Annotations only declare readOnlyHint=false, destructiveHint=false and openWorldHint=false; the description goes well beyond that by disclosing the return payload (approveUrl plus a code), the human-in-the-loop consent step in Warden, the guarantee that 'nothing changes until they do', and the request's 10-minute expiry. The one gap is what happens if the request expires or is declined, which matters for a grant that unlocks trading.

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 purpose and the dependency, then the return values, then the human instruction, then the polling step, then parameter semantics. Every sentence carries load-bearing information (URL, code, expiry, poll target, limits default); 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, and the description compensates by naming the returned approveUrl and code and telling the agent how to relay them. For a consent-gated session grant with a nested limits object, it covers timing, human interaction, default behavior and follow-up polling — enough to invoke and drive the flow correctly.

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 0%, so the description carries the load: it defines hours as session duration and limits as an optional override of the wallet's daily and per-trade limits in whole dollars, with an explicit keep-default behavior. Two weaknesses remain — the per-trade wording does not map cleanly to the nested humanApprovalAboveValue field, and agentName (max 40 chars) is never explained at all.

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 and resource ('Ask the wallet owner to turn on hands-free trading') and immediately links it to the sibling it unblocks ('so execute_swap works'). An agent can distinguish this from session_request_status or agent_permissions without reading any 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 frames the call as a prerequisite to execute_swap, names the follow-up tool to poll ('poll session_request_status every few seconds') and even supplies the cadence and the 10-minute window. It also gives an explicit omit rule for the optional limits parameter ('leave it out to keep them'), so both inclusion and exclusion paths are covered.

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