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

Beyond annotations (which only mark idempotent and non-read-only), the description discloses that the transcript is STORED and READABLE BY ANYONE with access to the application record, that scrubbing of credentials is best-effort, and that retry with the same key and turn indexes overwrites rather than duplicates. This is significant behavioral context for a mutation operation with privacy implications. No contradiction with annotations (idempotentHint matches the overwrite behavior).

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 paragraphs each serve a distinct purpose: purpose, privacy/safety, and operational mechanics. No filler sentences; the security caveat and idempotency detail are essential given the tool stores conversation data. Front-loads the action verb in the first sentence.

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 an 11-parameter, 5-required mutation tool with no output schema, the description is unusually complete: it covers user consent, privacy exposure, credential scrubbing limitations, ordering, and idempotency. The only minor omission is the response format, but that is not necessary for correct invocation. An agent has everything it needs to call safely.

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 covers 6 of 11 parameters (55%). The description adds crucial semantics for `sessionKey` (stable across calls for joining), `turns` (send in order, up to 50 per call), and idempotency (same key + turn indexes overwrites), which are not fully captured by the schema. It does not explain the remaining optional analytics parameters (agentName, modelName, releaseId, tokens), but those are self-explanatory and non-critical.

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 and resource: 'Attaches the conversation that produced this application to its record.' It also states the purpose (reference for maintainers and the approving reviewer), which clearly differentiates it from sibling tools like springroll.app.update or springroll.approval.submit. Title and description align on the same action.

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 explicitly commands 'Ask the user before calling this, every time,' establishing a hard precondition. It gives a when-not condition ('If the conversation contained a real secret, do not record it') and operational usage rules: send turns in order, keep sessionKey stable, and expect overwrite on retry. No sibling tool does the same job, so naming alternatives is unnecessary.

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