Recall: Open questions

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

Beyond the readOnly/idempotent/destructive annotations, the description discloses result provenance semantics that the agent cannot get anywhere else: which fields cite every result, the exact speaker values that make `owner` true, and the meaning of each `status_basis` value (user vs audit vs assistant vs verify). The closing prohibition against presenting model or audit output as the owner's decision is a concrete behavioral constraint that prevents a real error class.

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 what the tool returns, followed by citation/provenance detail and one safety rule. Every sentence carries information, though the middle sentence is dense with field enumerations that could be trimmed. No filler or restating of the title.

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?

With no output schema, the description carries the burden of describing results and does so well for the provenance fields (source_id, msg_id, ts, speaker, owner, status, status_basis). It omits ordering, pagination and what happens when limit truncates unresolved items, which matters for a recall/read tool with a default limit of 10. Read-only semantics are already covered by annotations.

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 100%, so all three parameters (limit, project, all_projects) are already documented in the schema, including defaults and the 1-50 range. The description adds no additional parameter meaning. Baseline 3 is correct when the schema does the full job.

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 states a specific resource and its scope precisely: unresolved questions plus recorded conflicts nobody settled, with cited evidence. An agent can distinguish this content type from recall_decisions or recall_forgotten_ideas by the returned subject matter. It never names a sibling family member explicitly, so differentiation is by inference rather than by routing language.

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?

Usage is implied by the content type (reach for this when you need open questions, not settled decisions), but there is no explicit when-to-use, when-not-to-use, or named alternative among recall_search, recall_ask, recall_decisions or recall_timeline. The one strong imperative present (never present an assistant/audit record as the owner's decision) is a correctness rule for downstream reporting, not a selection guideline.

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