replay_sessionRe-fire every captured request in a session against the real upstream. Returns one row per request with status, duration, and error, continuing on failure.
Re-fire every captured request in a session against the real upstream. Returns one row per attempted request with status, duration, and error if any. One failure does not abort the loop.
| Name | Required | Description | Default |
|---|---|---|---|
| id | Yes | Session id from list_sessions. |
Changes observed during successful MCP inspections.
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations cover safety profile (readOnlyHint=false, idempotentHint=false, openWorldHint=true), so the description's remaining burden is smaller. It adds genuinely useful behavioral facts beyond annotations: returns one row per attempt with status/duration/error, and crucially that one failure does not abort the loop. This error-tolerance detail is not derivable from annotations.
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 tight sentences, front-loaded with the action, then the return shape, then the failure-handling rule. Zero 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, so describing the return row (status, duration, error) is necessary and done. The non-aborting failure semantics fill a real gap. Missing only caveats like upstream reachability or whether it mutates captured state, which would fully close the loop.
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 100% and the single param 'id' is documented in the schema ('Session id from list_sessions'). Description adds no parameter detail. Baseline 3 is correct when the schema fully carries parameter semantics.
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?
Specific verb (re-fire) and resource (every captured request in a session) with scope stated precisely ('against the real upstream'). Distinguishes from replay_request sibling by operating on a whole session rather than a single request.
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?
Implies usage via 'captured request in a session' and 'real upstream', giving context, but never states when to use this over replay_request, nor any prerequisites (capture must exist, upstream reachable). No exclusions stated.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.