browser_resume

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

The annotations already declare destructiveHint=true, idempotentHint=false and openWorldHint=true; the description adds the useful fact that previously successful steps are not re-sent, which is real behavioral context. However, given the destructive flag it says nothing about what the re-sent (failed) steps may do, what happens if 'wait' expires, or whether resuming twice is safe — the risk side of a destructive tool is undisclosed. No contradiction with annotations, just an incomplete behavioral picture.

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?

One sentence, no padding, and the trigger condition is front-loaded before the behavioral clause. Slightly over-compressed wording costs a bit of readability but nothing is wasted.

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 destructive, non-idempotent tool with a nested 'observed' object and no output schema, the definition covers the happy path but omits failure handling: what happens when 'wait' elapses without user input, and how the 'observed' applied/not_applied map should be used to recover a partially-applied step. The schema covers 'observed' mechanics, which keeps this from being lower, but the operational gaps remain.

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 coverage is only 50%, and the description specifically compensates for the two undocumented parameters: it tells the agent that 'call' and 'resume_token' are values taken from a previous call's result, not agent-invented inputs. The remaining 'wait' and 'observed' parameters are fully documented in the schema itself, so the combined coverage is complete.

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 uses a specific verb ('Continue') on a specific resource ('a call') and pins the trigger to a prior result carrying 'call' and 'resume_token' from a needs_user stop, batch, parallel or workflow. That is enough to distinguish it from every sibling, none of which resume an interrupted call. It stops short of 5 because the compressed phrasing ('nothing that ran ok is sent again') takes a second read to parse.

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 states exactly when to use the tool: when a prior result contains 'call' and 'resume_token' from one of four interruption types. No alternatives or when-not conditions are named, but no sibling competes for this job, so the trigger alone is sufficient.

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