record_experience

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

With no annotations, the description carries the full burden and delivers real behavioral content: idempotency ('exact retries do not duplicate evidence'), provenance caveats (caller-reported, not server-verified), and correction semantics ('invalidates the old review and retracts its effects atomically'). It does not disclose what happens on a revision mismatch or other failure modes, so it is not exhaustive.

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?

Five dense sentences, all front-loaded with the purpose before mechanics, and every sentence carries information (revision protocol, exclusion, idempotency, provenance, atomic correction). The 'review-owned candidates avoid that duplicate prior' phrasing is jargon-heavy and slightly opaque, but nothing is 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?

For a nested-object mutation tool with no annotations and no output schema, the description covers the concurrency contract, idempotency, provenance trust level, and correction side effects. The remaining gaps are return-value shape and conflict/error behavior on stale revisions.

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%, and while the description explains expected_revision well (0 to create, returned revision to correct) and hints at id as a stable incident key, the seven required nested experience fields (project, context, action, outcome, source, evidence) get no explanation. The enums are self-describing, but the description only partially compensates for the coverage gap.

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 with scope: 'Record one observed action and outcome for incremental LLM review.' The 'one observed action and outcome' framing plus the correction-vs-create semantics clearly separate it from siblings like review_experience, pending_experiences, and insert_belief.

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?

Gives concrete operational guidance: stable UUID for the same incident, expected_revision=0 to create, returned revision to correct. It also states an explicit exclusion ('Do not also insert a standalone conclusion from this same incident'), which routes the agent away from a duplicating sibling. It stops short of naming that sibling tool by name, so it is strong but not maximal.

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