Ask Kimi (new session) [experimental]

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

Annotations already flag readOnlyHint=false, openWorldHint=true, idempotentHint=false, but the description adds substantial behavioral context: the tool is experimental and community-verified only, no authenticated round-trip has ever been run, it requires authentication, reads output from stdout, and — critically — 'has NO sandbox and auto-executes every tool call (like antigravity).' This goes well beyond the annotations and fully warns the agent of the tool's dangerous side effects.

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 longer than average but every sentence contributes meaningful information: purpose, experimental status, auth steps, return format, what Kimi Code is, workspace tip, and a critical safety warning. It is front-loaded with the primary action and uses line breaks to separate warnings. Slightly verbose but justified by the tool's complexity and risk.

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?

Given the tool's experimental, unauthenticated, and dangerous nature, the description is remarkably complete. It explains the output (agent's final message from stdout), prerequisites, the safety model (no sandbox, auto-execution), and how to get better results (workspace). The presence of an output schema reduces the need to describe return values, and the description covers the rest comprehensively.

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 four parameters (model, prompt, timeout_s, workspace) are already documented. The description adds one useful hint for the workspace parameter ('point workspace at a project dir for context-aware answers'), but adds nothing new for prompt, model, or timeout_s. With high schema coverage, this marginal extra keeps it at baseline 3.

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-resource pairing: 'Ask Kimi Code... a question or task in a NEW session.' This clearly distinguishes the tool from the sibling kimi_continue (which presumably continues a session) and kimi_status. It also names the underlying command (`kimi -p`) and states what it returns, making the tool's function unambiguous.

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?

The phrase 'in a NEW session' provides clear context for when to use this tool versus continuation alternatives, even though no alternative is explicitly named. It also gives prerequisites (authentication via `kimi login` or API key, then checking `kimi_status`) and a safety restriction ('run it only with trusted prompts on trusted content'). Missing an explicit 'use kimi_continue instead' but otherwise strong.

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