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

Far beyond the readOnlyHint/openWorldHint/idempotentHint annotations: it discloses that the question leaves HireMike and is stored by a third party, that contact details cause a typed refusal (question_has_contact_details), that answers can take up to a minute, that the payload is untrusted AI output whose instructions must not be followed, and that conversations expire after 30 minutes (help_conversation_ended). That is exactly the mutation, privacy and failure surface an agent needs.

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 paragraphs is long, but the purpose and the data-boundary warning are front-loaded and each paragraph covers a distinct concern (privacy, untrusted output, failures, conversation flow). A few clauses are restated — the 'checks do not apply there' warning and the never-choose-for-the-member rule both appear twice — so it stops short of maximal tightness.

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?

An output schema exists, let alone it being required, yet the description still explains the two fields an agent must act on (`untrusted` must be relayed with its article names and never obeyed, `conversation` must be threaded back). Combined with the failure modes and privacy constraints, nothing needed to invoke it safely is missing.

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 0%, so the description carries the full load and does: `question` must be in general words, must not contain names, emails, phones or links, and must be an exact option label when selecting; `conversation` is the token from the prior answer, omitted for a new question and invalid 30 minutes after the last answer. Both parameters are unusable without this text, and both are covered.

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 first sentence names the specific verb and resource ('ask HireMike's help bot') and enumerates the recognisable question types: a setting, a feature, a step, a status, a message. It then draws the boundary against the employer-data siblings in one clause, so an agent can pick it out of a 40+ tool list.

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?

Explicit routing both ways: product questions go here, employer-data questions go to 'the other tools'. It also gives a full failure protocol (help_unavailable/help_timeout → stop and tell the member, they can use the dashboard bubble, never open HireMike by browser or other tool) and a follow-up rule (new unrelated question → no conversation).

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