ask-kimi
Destructive
Send coding tasks or questions to Kimi Code CLI with file analysis and session support. Resolve problems via natural language, including file references and safety controls.
Instructions
Execute Kimi Code CLI with file analysis (@syntax), session resume, MCP passthrough, and safety controls.
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| cd | No | Working directory | |
| afk | No | --afk: auto-approve in away-from-keyboard mode (print mode implies this). | |
| plan | No | Start in plan mode (--plan). For richer planning, prefer the plan-kimi tool. | |
| yolo | No | ⚠️ --yolo: auto-approve everything (dangerous, full file access). | |
| agent | No | Builtin agent spec to use (--agent). "default" or "okabe". | |
| model | No | Optional model override. Known: kimi-for-coding, kimi-k2-6, kimi-k2-6-turbo, kimi-k2-thinking-turbo, kimi-k2-5, kimi-latest. If omitted, uses your Kimi CLI default (~/.kimi/config.toml). | |
| quiet | No | Use --quiet (= --print --output-format text --final-message-only). | |
| config | No | Configuration overrides — string (JSON/TOML) or object passed via --config. | |
| dryRun | No | Return the resolved CLI argv as JSON without spawning Kimi. | |
| prompt | Yes | Task or question. Use @ to include files (e.g., '@largefile.ts explain'). | |
| addDirs | No | Additional writable directories (--add-dir, repeatable). | |
| timeout | No | Maximum execution time in milliseconds (optional) | |
| thinking | No | Toggle thinking mode (overrides default_thinking from config). | |
| agentFile | No | Custom agent spec file (--agent-file). | |
| finalOnly | No | Add --final-message-only — skip intermediate tool-call output. Default: true. | |
| sessionId | No | Internal session ID for conversation continuity (workspace-isolated). | |
| configFile | No | Path to a config file (TOML or JSON). | |
| skillsDirs | No | Custom skills directories (--skills-dir, repeatable). Overrides default discovery. | |
| workingDir | No | Working directory for execution (alias of cd) | |
| outputFormat | No | Kimi --output-format. "text" (default for clean responses) or "stream-json" (JSONL with assistant + tool messages). | |
| resetSession | No | Clear session context before execution. Starts fresh conversation. | |
| responseMode | No | Response verbosity: "clean" returns only the final answer (default), "full" includes tool calls and stderr execution log. | clean |
| mcpConfigFile | No | Path to an MCP config file with `mcpServers` (passed via --mcp-config-file). | |
| continueLatest | No | Use --continue to resume the most recent Kimi session in cwd. | |
| includeMetadata | No | Include tool-call metadata in response (only meaningful with stream-json). | |
| maxStepsPerTurn | No | Cap steps in one turn (--max-steps-per-turn). | |
| maxRalphIterations | No | Extra autonomous iterations after the first turn (--max-ralph-iterations, -1 = unlimited). |