ask_kimi

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

No annotations are present, so the description carries the full burden — and it delivers: single-turn mode, sandboxed pure-text reasoning with no filesystem/tool access, the 131k-byte argv cap with refusal behavior, prerequisite `kimi` CLI + `kimi login` with error states, and explicit refusal boundaries (offensive security, biology/medicine refused; neuroscience, cognitive science, AI/ML, CS in-scope). This is unusually thorough behavioral disclosure.

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?

The description is long but front-loaded: purpose first, then routing preference, then caveats, sandbox, prerequisites, refusals, alternatives — each sentence earns its place given the zero-annotation environment. However, the dense em-dash and parenthetical asides make parsing harder than necessary, so it falls just short of top marks.

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 annotations and no output schema, the description must be self-sufficient, and it covers nearly everything: purpose, alternatives, prerequisites, error states, size limits, refusal boundaries, and cost implications. It does not explicitly describe the return value format, though 'sandboxed to pure text reasoning' strongly implies a text answer — a minor gap for an otherwise complete definition.

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%, so the baseline is 3. The description adds meaning beyond the schema by explaining why `context` matters ('the model has NO filesystem or tool access, so put the real code in `context`') and by tying the `question` parameter to the argv size cap. This goes one step beyond the schema's own documentation.

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 names a specific verb ('Ask'), a specific resource (Moonshot's Kimi model kimi-code/k3 via the local `kimi` CLI), and the scope (SOFTWARE/ENGINEERING work: code structure, functionality, data/control flow, architecture). It distinguises from sibling tools by adding 'on its own, independent of Fable' and explicitly contrasts itself with `ask_atlas`, `ask`, `ask_m3`, and `ask_council'.

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 explicit when-to-use: 'PREFER THIS over `ask_atlas`... whenever the `kimi` binary is installed', with the billing rationale. Gives when-not-to-use: prompts over the 131k-byte argv cap should be routed to `ask_atlas`. Also names the siblings for other purposes: 'Use `ask` for Fable, `ask_m3` for MiniMax, or `ask_council` to ask several'.

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