Skip to main content
Glama
mkXultra

ai-cli-mcp

by mkXultra
OverviewSchemaSourceRelated ServersDiscussionsAdmin
TypeScript
Local

run

Launch an AI coding agent as a background process and receive a PID instantly, enabling asynchronous monitoring and result retrieval.

Instructions

AI Agent Runner: Starts a Claude, Codex, Antigravity (Gemini), OpenCode, Grok, or Pi CLI process in the background and returns a PID immediately. Use list_processes and get_result to monitor progress.

• File ops: Create, read, (fuzzy) edit, move, copy, delete, list files, analyze/ocr images, file content analysis • Code: Generate / analyse / refactor / fix • Git: Stage ▸ commit ▸ push ▸ tag (any workflow) • Terminal: Run any CLI cmd or open URLs • Web search + summarise content on-the-fly • Multi-step workflows & GitHub integration

IMPORTANT: This tool now returns immediately with a PID. Use other tools to check status and get results.

Supported models: "claude-ultra", "codex-ultra", "gemini-ultra", "sol", "luna", "sonnet", "sonnet[1m]", "opus", "opusplan", "fable", "haiku", "gpt-6-astra", "gpt-6-sol", "gpt-6-luna", "gpt-5.4", "gpt-5.6-sol", "gpt-5.6-terra", "gpt-5.6-luna", "gpt-5.5", "gpt-5.4-mini", "gpt-5.3-codex", "gpt-5.3-codex-spark", "gpt-5.2", "gemini-3.8-flash-high", "gemini-3.8-flash-medium", "gemini-3.8-flash-low", "gemini-3.7-flash-high", "gemini-3.7-flash-medium", "gemini-3.7-flash-low", "gemini-3.6-flash-high", "gemini-3.6-flash-medium", "gemini-3.6-flash-low", "gemini-3.1-pro-high", "gemini-3.1-pro-low", "grok", "grok-4.6", "grok-4.5", "opencode", "pi", "oc-<provider/model>", "pi-<provider/model>"

Prompt input: You must provide EITHER prompt (string) OR prompt_file (file path), but not both.

Prompt tips

  1. Be concise, explicit & step-by-step for complex tasks.

  2. Check process status with list_processes

  3. Get results with get_result using the returned PID

  4. Kill long-running processes with kill_process if needed

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
modelNoThe model to use. Aliases (including user config): "claude-ultra" (opus; auto max reasoning), "codex-ultra" (gpt-6-astra; auto ultra reasoning), "gemini-ultra" (gemini-3.8-flash-high), "sol" (gpt-6-sol), "luna" (gpt-6-luna). An explicit reasoning_effort overrides the alias default. Standard: "sonnet", "sonnet[1m]", "opus", "opusplan", "fable", "haiku", "gpt-6-astra", "gpt-6-sol", "gpt-6-luna", "gpt-5.4", "gpt-5.6-sol", "gpt-5.6-terra", "gpt-5.6-luna", "gpt-5.5", "gpt-5.4-mini", "gpt-5.3-codex", "gpt-5.3-codex-spark", "gpt-5.2", "gemini-3.8-flash-high", "gemini-3.8-flash-medium", "gemini-3.8-flash-low", "gemini-3.7-flash-high", "gemini-3.7-flash-medium", "gemini-3.7-flash-low", "gemini-3.6-flash-high", "gemini-3.6-flash-medium", "gemini-3.6-flash-low", "gemini-3.1-pro-high", "gemini-3.1-pro-low", "grok", "grok-4.6", "grok-4.5", "opencode", "pi". Gemini models run through Antigravity CLI (agy); use agy models for current availability. Model effort suffixes must match reasoning_effort when supplied. Fable may require usage credits. Grok accepts native grok-* model names; grok uses its CLI-configured default unless a user alias named grok exists; that alias retains precedence. Native model names such as grok-4.6 select the Grok backend. Omitted effort uses the CLI default; grok and unknown Grok models accept low/medium/high, grok-4.6 also accepts xhigh. OpenCode accepts explicit dynamic models using "oc-<provider/model>". Pi accepts its configured default with "pi" or an explicit dynamic model using "pi-<provider/model>"; discover current choices with "pi --list-models".
promptNoThe detailed natural language prompt for the agent to execute. Either this or prompt_file is required.
session_idNoOptional session ID to resume a previous session. Supported for Claude, Codex, Gemini, OpenCode, Grok, and Pi. Antigravity resumes its conversation_id via --conversation; old Gemini CLI sessions cannot be resumed. Grok resumes via --resume, preserving the session ID. OpenCode and Pi resume in place via --session and may also be combined with explicit model selection.
workFolderYesThe working directory for the agent execution. Must be an absolute path.
prompt_fileNoPath to a file containing the prompt. Either this or prompt is required. Must be an absolute path or relative to workFolder.
reasoning_effortNoReasoning control for Claude, Codex, Antigravity (Gemini), Grok, and Pi. Claude uses --effort with "low", "medium", "high", "xhigh", "max". Codex uses model_reasoning_effort with "low", "medium", "high", "xhigh"; GPT-6 Astra/Sol and GPT-5.6 Sol/Terra also support "max" and "ultra", while GPT-6 Luna and GPT-5.6 Luna support "max". Grok: grok-4.6=low/medium/high/xhigh; grok-4.5, grok (configured default), and other grok-* models=low/medium/high. Pi uses --thinking with off/minimal/low/medium/high/xhigh/max. Omitted effort uses the CLI default. Antigravity uses --effort low/medium/high; it must match any effort suffix in the model name. OpenCode does not support reasoning_effort in this integration.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed3 schema fields changedv3.0.0
    • changedInput schema / properties / model / description
      Previous value: -"The model to use. Aliases (including user config): \"claude-ultra\" (opus; auto max reasoning), \"codex-ultra\" (gpt-6-astra; auto ultra reasoning), \"gemini-ultra\" (gemini-3.8-flash-high). An explicit reasoning_effort overrides the alias default. Standard: \"sonnet\", \"sonnet[1m]\", \"opus\", \"opusplan\", \"fable\", \"haiku\", \"gpt-6-astra\", \"gpt-5.4\", \"gpt-5.6-sol\", \"gpt-5.6-terra\", \"gpt-5.6-luna\", \"gpt-5.5\", \"gpt-5.4-mini\", \"gpt-5.3-codex\", \"gpt-5.3-codex-spark\", \"gpt-5.2\", \"gemini-3.8-flash-high\", \"gemini-3.8-flash-medium\", \"gemini-3.8-flash-low\", \"gemini-3.7-flash-high\", \"gemini-3.7-flash-medium\", \"gemini-3.7-flash-low\", \"gemini-3.6-flash-high\", \"gemini-3.6-flash-medium\", \"gemini-3.6-flash-low\", \"gemini-3.1-pro-high\", \"gemini-3.1-pro-low\", \"forge\", \"grok\", \"grok-4.6\", \"grok-4.5\", \"opencode\". Gemini models run through Antigravity CLI (agy); use agy models for current availability. Model effort suffixes must match reasoning_effort when supplied. Fable may require usage credits. Grok accepts native grok-* model names; grok uses its CLI-configured default unless a user alias named grok exists; that alias retains precedence. Native model names such as grok-4.6 select the Grok backend. Omitted effort uses the CLI default; grok and unknown Grok models accept low/medium/high, grok-4.6 also accepts xhigh. OpenCode also accepts explicit dynamic models using \"oc-<provider/model>\". \"forge\" is a provider key, not a Forge model family selector."New value: +"The model to use. Aliases (including user config): \"claude-ultra\" (opus; auto max reasoning), \"codex-ultra\" (gpt-6-astra; auto ultra reasoning), \"gemini-ultra\" (gemini-3.8-flash-high), \"sol\" (gpt-6-sol), \"luna\" (gpt-6-luna). An explicit reasoning_effort overrides the alias default. Standard: \"sonnet\", \"sonnet[1m]\", \"opus\", \"opusplan\", \"fable\", \"haiku\", \"gpt-6-astra\", \"gpt-6-sol\", \"gpt-6-luna\", \"gpt-5.4\", \"gpt-5.6-sol\", \"gpt-5.6-terra\", \"gpt-5.6-luna\", \"gpt-5.5\", \"gpt-5.4-mini\", \"gpt-5.3-codex\", \"gpt-5.3-codex-spark\", \"gpt-5.2\", \"gemini-3.8-flash-high\", \"gemini-3.8-flash-medium\", \"gemini-3.8-flash-low\", \"gemini-3.7-flash-high\", \"gemini-3.7-flash-medium\", \"gemini-3.7-flash-low\", \"gemini-3.6-flash-high\", \"gemini-3.6-flash-medium\", \"gemini-3.6-flash-low\", \"gemini-3.1-pro-high\", \"gemini-3.1-pro-low\", \"grok\", \"grok-4.6\", \"grok-4.5\", \"opencode\", \"pi\". Gemini models run through Antigravity CLI (agy); use agy models for current availability. Model effort suffixes must match reasoning_effort when supplied. Fable may require usage credits. Grok accepts native grok-* model names; grok uses its CLI-configured default unless a user alias named grok exists; that alias retains precedence. Native model names such as grok-4.6 select the Grok backend. Omitted effort uses the CLI default; grok and unknown Grok models accept low/medium/high, grok-4.6 also accepts xhigh. OpenCode accepts explicit dynamic models using \"oc-<provider/model>\". Pi accepts its configured default with \"pi\" or an explicit dynamic model using \"pi-<provider/model>\"; discover current choices with \"pi --list-models\"."
    • changedInput schema / properties / reasoning_effort / description
      Previous value: -"Reasoning control for Claude, Codex, Antigravity (Gemini), and Grok. Claude uses --effort with \"low\", \"medium\", \"high\", \"xhigh\", \"max\". Codex uses model_reasoning_effort with \"low\", \"medium\", \"high\", \"xhigh\"; GPT-6 Astra and GPT-5.6 Sol/Terra also support \"max\" and \"ultra\", while GPT-5.6 Luna supports \"max\". Grok: grok-4.6=low/medium/high/xhigh; grok-4.5, grok (configured default), and other grok-* models=low/medium/high. Omitted effort uses the CLI default. Grok never accepts max/ultra. Antigravity uses --effort low/medium/high; it must match any effort suffix in the model name. Forge and OpenCode do not support reasoning_effort in this integration."New value: +"Reasoning control for Claude, Codex, Antigravity (Gemini), Grok, and Pi. Claude uses --effort with \"low\", \"medium\", \"high\", \"xhigh\", \"max\". Codex uses model_reasoning_effort with \"low\", \"medium\", \"high\", \"xhigh\"; GPT-6 Astra/Sol and GPT-5.6 Sol/Terra also support \"max\" and \"ultra\", while GPT-6 Luna and GPT-5.6 Luna support \"max\". Grok: grok-4.6=low/medium/high/xhigh; grok-4.5, grok (configured default), and other grok-* models=low/medium/high. Pi uses --thinking with off/minimal/low/medium/high/xhigh/max. Omitted effort uses the CLI default. Antigravity uses --effort low/medium/high; it must match any effort suffix in the model name. OpenCode does not support reasoning_effort in this integration."
    • changedInput schema / properties / session_id / description
      Previous value: -"Optional session ID to resume a previous session. Supported for Claude, Codex, Gemini, Forge, OpenCode, and Grok. Antigravity resumes its conversation_id via --conversation; old Gemini CLI sessions cannot be resumed. Grok resumes via --resume, preserving the session ID. OpenCode resumes in-place via --session and may also be combined with explicit oc-<provider/model> selection."New value: +"Optional session ID to resume a previous session. Supported for Claude, Codex, Gemini, OpenCode, Grok, and Pi. Antigravity resumes its conversation_id via --conversation; old Gemini CLI sessions cannot be resumed. Grok resumes via --resume, preserving the session ID. OpenCode and Pi resume in place via --session and may also be combined with explicit model selection."
  2. Changed3 schema fields changedv2.27.0
    • changedInput schema / properties / model / description
      Previous value: -"The model to use. Aliases (including user config): \"claude-ultra\" (opus; auto max reasoning), \"codex-ultra\" (gpt-6-astra; auto ultra reasoning), \"gemini-ultra\" (gemini-3.1-pro-preview). An explicit reasoning_effort overrides the alias default. Standard: \"sonnet\", \"sonnet[1m]\", \"opus\", \"opusplan\", \"fable\", \"haiku\", \"gpt-6-astra\", \"gpt-5.4\", \"gpt-5.6-sol\", \"gpt-5.6-terra\", \"gpt-5.6-luna\", \"gpt-5.5\", \"gpt-5.4-mini\", \"gpt-5.3-codex\", \"gpt-5.3-codex-spark\", \"gpt-5.2\", \"gemini-2.5-pro\", \"gemini-2.5-flash\", \"gemini-3.1-pro-preview\", \"gemini-3-pro-preview\", \"gemini-3-flash-preview\", \"forge\", \"grok\", \"grok-4.6\", \"grok-4.5\", \"opencode\". Fable may require usage credits. Grok accepts native grok-* model names; grok uses its CLI-configured default unless a user alias named grok exists; that alias retains precedence. Native model names such as grok-4.6 select the Grok backend. Omitted effort uses the CLI default; grok and unknown Grok models accept low/medium/high, grok-4.6 also accepts xhigh. OpenCode also accepts explicit dynamic models using \"oc-<provider/model>\". \"forge\" is a provider key, not a Forge model family selector."New value: +"The model to use. Aliases (including user config): \"claude-ultra\" (opus; auto max reasoning), \"codex-ultra\" (gpt-6-astra; auto ultra reasoning), \"gemini-ultra\" (gemini-3.8-flash-high). An explicit reasoning_effort overrides the alias default. Standard: \"sonnet\", \"sonnet[1m]\", \"opus\", \"opusplan\", \"fable\", \"haiku\", \"gpt-6-astra\", \"gpt-5.4\", \"gpt-5.6-sol\", \"gpt-5.6-terra\", \"gpt-5.6-luna\", \"gpt-5.5\", \"gpt-5.4-mini\", \"gpt-5.3-codex\", \"gpt-5.3-codex-spark\", \"gpt-5.2\", \"gemini-3.8-flash-high\", \"gemini-3.8-flash-medium\", \"gemini-3.8-flash-low\", \"gemini-3.7-flash-high\", \"gemini-3.7-flash-medium\", \"gemini-3.7-flash-low\", \"gemini-3.6-flash-high\", \"gemini-3.6-flash-medium\", \"gemini-3.6-flash-low\", \"gemini-3.1-pro-high\", \"gemini-3.1-pro-low\", \"forge\", \"grok\", \"grok-4.6\", \"grok-4.5\", \"opencode\". Gemini models run through Antigravity CLI (agy); use agy models for current availability. Model effort suffixes must match reasoning_effort when supplied. Fable may require usage credits. Grok accepts native grok-* model names; grok uses its CLI-configured default unless a user alias named grok exists; that alias retains precedence. Native model names such as grok-4.6 select the Grok backend. Omitted effort uses the CLI default; grok and unknown Grok models accept low/medium/high, grok-4.6 also accepts xhigh. OpenCode also accepts explicit dynamic models using \"oc-<provider/model>\". \"forge\" is a provider key, not a Forge model family selector."
    • changedInput schema / properties / reasoning_effort / description
      Previous value: -"Reasoning control for Claude, Codex, and Grok. Claude uses --effort with \"low\", \"medium\", \"high\", \"xhigh\", \"max\". Codex uses model_reasoning_effort with \"low\", \"medium\", \"high\", \"xhigh\"; GPT-6 Astra and GPT-5.6 Sol/Terra also support \"max\" and \"ultra\", while GPT-5.6 Luna supports \"max\". Grok: grok-4.6=low/medium/high/xhigh; grok-4.5, grok (configured default), and other grok-* models=low/medium/high. Omitted effort uses the CLI default. Grok never accepts max/ultra. Gemini, Forge, and OpenCode do not support reasoning_effort in this integration."New value: +"Reasoning control for Claude, Codex, Antigravity (Gemini), and Grok. Claude uses --effort with \"low\", \"medium\", \"high\", \"xhigh\", \"max\". Codex uses model_reasoning_effort with \"low\", \"medium\", \"high\", \"xhigh\"; GPT-6 Astra and GPT-5.6 Sol/Terra also support \"max\" and \"ultra\", while GPT-5.6 Luna supports \"max\". Grok: grok-4.6=low/medium/high/xhigh; grok-4.5, grok (configured default), and other grok-* models=low/medium/high. Omitted effort uses the CLI default. Grok never accepts max/ultra. Antigravity uses --effort low/medium/high; it must match any effort suffix in the model name. Forge and OpenCode do not support reasoning_effort in this integration."
    • changedInput schema / properties / session_id / description
      Previous value: -"Optional session ID to resume a previous session. Supported for Claude, Codex, Gemini, Forge, OpenCode, and Grok. Grok resumes via --resume, preserving the session ID. OpenCode resumes in-place via --session and may also be combined with explicit oc-<provider/model> selection."New value: +"Optional session ID to resume a previous session. Supported for Claude, Codex, Gemini, Forge, OpenCode, and Grok. Antigravity resumes its conversation_id via --conversation; old Gemini CLI sessions cannot be resumed. Grok resumes via --resume, preserving the session ID. OpenCode resumes in-place via --session and may also be combined with explicit oc-<provider/model> selection."
  3. Changed3 schema fields changedv2.26.0
    • changedInput schema / properties / model / description
      Previous value: -"The model to use. Aliases (including user config): \"claude-ultra\" (opus; auto max reasoning), \"codex-ultra\" (gpt-6-astra; auto ultra reasoning), \"gemini-ultra\" (gemini-3.1-pro-preview). An explicit reasoning_effort overrides the alias default. Standard: \"sonnet\", \"sonnet[1m]\", \"opus\", \"opusplan\", \"fable\", \"haiku\", \"gpt-6-astra\", \"gpt-5.4\", \"gpt-5.6-sol\", \"gpt-5.6-terra\", \"gpt-5.6-luna\", \"gpt-5.5\", \"gpt-5.4-mini\", \"gpt-5.3-codex\", \"gpt-5.3-codex-spark\", \"gpt-5.2\", \"gemini-2.5-pro\", \"gemini-2.5-flash\", \"gemini-3.1-pro-preview\", \"gemini-3-pro-preview\", \"gemini-3-flash-preview\", \"forge\", \"opencode\". Fable may require usage credits. OpenCode also accepts explicit dynamic models using \"oc-<provider/model>\". \"forge\" is a provider key, not a Forge model family selector."New value: +"The model to use. Aliases (including user config): \"claude-ultra\" (opus; auto max reasoning), \"codex-ultra\" (gpt-6-astra; auto ultra reasoning), \"gemini-ultra\" (gemini-3.1-pro-preview). An explicit reasoning_effort overrides the alias default. Standard: \"sonnet\", \"sonnet[1m]\", \"opus\", \"opusplan\", \"fable\", \"haiku\", \"gpt-6-astra\", \"gpt-5.4\", \"gpt-5.6-sol\", \"gpt-5.6-terra\", \"gpt-5.6-luna\", \"gpt-5.5\", \"gpt-5.4-mini\", \"gpt-5.3-codex\", \"gpt-5.3-codex-spark\", \"gpt-5.2\", \"gemini-2.5-pro\", \"gemini-2.5-flash\", \"gemini-3.1-pro-preview\", \"gemini-3-pro-preview\", \"gemini-3-flash-preview\", \"forge\", \"grok\", \"grok-4.6\", \"grok-4.5\", \"opencode\". Fable may require usage credits. Grok accepts native grok-* model names; grok uses its CLI-configured default unless a user alias named grok exists; that alias retains precedence. Native model names such as grok-4.6 select the Grok backend. Omitted effort uses the CLI default; grok and unknown Grok models accept low/medium/high, grok-4.6 also accepts xhigh. OpenCode also accepts explicit dynamic models using \"oc-<provider/model>\". \"forge\" is a provider key, not a Forge model family selector."
    • changedInput schema / properties / reasoning_effort / description
      Previous value: -"Reasoning control for Claude and Codex. Claude uses --effort with \"low\", \"medium\", \"high\", \"xhigh\", \"max\". Codex uses model_reasoning_effort with \"low\", \"medium\", \"high\", \"xhigh\"; GPT-6 Astra and GPT-5.6 Sol/Terra also support \"max\" and \"ultra\", while GPT-5.6 Luna supports \"max\". Gemini, Forge, and OpenCode do not support reasoning_effort in this integration."New value: +"Reasoning control for Claude, Codex, and Grok. Claude uses --effort with \"low\", \"medium\", \"high\", \"xhigh\", \"max\". Codex uses model_reasoning_effort with \"low\", \"medium\", \"high\", \"xhigh\"; GPT-6 Astra and GPT-5.6 Sol/Terra also support \"max\" and \"ultra\", while GPT-5.6 Luna supports \"max\". Grok: grok-4.6=low/medium/high/xhigh; grok-4.5, grok (configured default), and other grok-* models=low/medium/high. Omitted effort uses the CLI default. Grok never accepts max/ultra. Gemini, Forge, and OpenCode do not support reasoning_effort in this integration."
    • changedInput schema / properties / session_id / description
      Previous value: -"Optional session ID to resume a previous session. Supported for Claude, Codex, Gemini, Forge, and OpenCode. OpenCode resumes in-place via --session and may also be combined with explicit oc-<provider/model> selection."New value: +"Optional session ID to resume a previous session. Supported for Claude, Codex, Gemini, Forge, OpenCode, and Grok. Grok resumes via --resume, preserving the session ID. OpenCode resumes in-place via --session and may also be combined with explicit oc-<provider/model> selection."
  4. Changed1 schema field changedv2.24.0
    • changedInput schema / properties / model / description
      Previous value: -"The model to use. Aliases: \"claude-ultra\" (Opus with auto max effort; does not select Fable), \"codex-ultra\" (auto ultra reasoning), \"gemini-ultra\". Standard: \"sonnet\", \"sonnet[1m]\", \"opus\", \"opusplan\", \"fable\", \"haiku\", \"gpt-6-astra\", \"gpt-5.4\", \"gpt-5.6-sol\", \"gpt-5.6-terra\", \"gpt-5.6-luna\", \"gpt-5.5\", \"gpt-5.4-mini\", \"gpt-5.3-codex\", \"gpt-5.3-codex-spark\", \"gpt-5.2\", \"gemini-2.5-pro\", \"gemini-2.5-flash\", \"gemini-3.1-pro-preview\", \"gemini-3-pro-preview\", \"gemini-3-flash-preview\", \"forge\", \"opencode\". Fable is an explicit selection and may require usage credits. OpenCode also accepts explicit dynamic models using \"oc-<provider/model>\". \"forge\" is a provider key, not a Forge model family selector."New value: +"The model to use. Aliases (including user config): \"claude-ultra\" (opus; auto max reasoning), \"codex-ultra\" (gpt-6-astra; auto ultra reasoning), \"gemini-ultra\" (gemini-3.1-pro-preview). An explicit reasoning_effort overrides the alias default. Standard: \"sonnet\", \"sonnet[1m]\", \"opus\", \"opusplan\", \"fable\", \"haiku\", \"gpt-6-astra\", \"gpt-5.4\", \"gpt-5.6-sol\", \"gpt-5.6-terra\", \"gpt-5.6-luna\", \"gpt-5.5\", \"gpt-5.4-mini\", \"gpt-5.3-codex\", \"gpt-5.3-codex-spark\", \"gpt-5.2\", \"gemini-2.5-pro\", \"gemini-2.5-flash\", \"gemini-3.1-pro-preview\", \"gemini-3-pro-preview\", \"gemini-3-flash-preview\", \"forge\", \"opencode\". Fable may require usage credits. OpenCode also accepts explicit dynamic models using \"oc-<provider/model>\". \"forge\" is a provider key, not a Forge model family selector."
  5. Changed2 schema fields changedv2.23.0
    • changedInput schema / properties / model / description
      Previous value: -"The model to use. Aliases: \"claude-ultra\" (Opus with auto max effort; does not select Fable), \"codex-ultra\" (auto ultra reasoning), \"gemini-ultra\". Standard: \"sonnet\", \"sonnet[1m]\", \"opus\", \"opusplan\", \"fable\", \"haiku\", \"gpt-5.4\", \"gpt-5.6-sol\", \"gpt-5.6-terra\", \"gpt-5.6-luna\", \"gpt-5.5\", \"gpt-5.4-mini\", \"gpt-5.3-codex\", \"gpt-5.3-codex-spark\", \"gpt-5.2\", \"gemini-2.5-pro\", \"gemini-2.5-flash\", \"gemini-3.1-pro-preview\", \"gemini-3-pro-preview\", \"gemini-3-flash-preview\", \"forge\", \"opencode\". Fable is an explicit selection and may require usage credits. OpenCode also accepts explicit dynamic models using \"oc-<provider/model>\". \"forge\" is a provider key, not a Forge model family selector."New value: +"The model to use. Aliases: \"claude-ultra\" (Opus with auto max effort; does not select Fable), \"codex-ultra\" (auto ultra reasoning), \"gemini-ultra\". Standard: \"sonnet\", \"sonnet[1m]\", \"opus\", \"opusplan\", \"fable\", \"haiku\", \"gpt-6-astra\", \"gpt-5.4\", \"gpt-5.6-sol\", \"gpt-5.6-terra\", \"gpt-5.6-luna\", \"gpt-5.5\", \"gpt-5.4-mini\", \"gpt-5.3-codex\", \"gpt-5.3-codex-spark\", \"gpt-5.2\", \"gemini-2.5-pro\", \"gemini-2.5-flash\", \"gemini-3.1-pro-preview\", \"gemini-3-pro-preview\", \"gemini-3-flash-preview\", \"forge\", \"opencode\". Fable is an explicit selection and may require usage credits. OpenCode also accepts explicit dynamic models using \"oc-<provider/model>\". \"forge\" is a provider key, not a Forge model family selector."
    • changedInput schema / properties / reasoning_effort / description
      Previous value: -"Reasoning control for Claude and Codex. Claude uses --effort with \"low\", \"medium\", \"high\", \"xhigh\", \"max\". Codex uses model_reasoning_effort with \"low\", \"medium\", \"high\", \"xhigh\"; GPT-5.6 Sol and Terra also support \"max\" and \"ultra\", while Luna supports \"max\". Gemini, Forge, and OpenCode do not support reasoning_effort in this integration."New value: +"Reasoning control for Claude and Codex. Claude uses --effort with \"low\", \"medium\", \"high\", \"xhigh\", \"max\". Codex uses model_reasoning_effort with \"low\", \"medium\", \"high\", \"xhigh\"; GPT-6 Astra and GPT-5.6 Sol/Terra also support \"max\" and \"ultra\", while GPT-5.6 Luna supports \"max\". Gemini, Forge, and OpenCode do not support reasoning_effort in this integration."
  6. Changed2 schema fields changedv2.22.0
    • changedInput schema / properties / model / description
      Previous value: -"The model to use. Aliases: \"claude-ultra\" (auto max effort), \"codex-ultra\" (auto xhigh reasoning), \"gemini-ultra\". Standard: \"sonnet\", \"sonnet[1m]\", \"opus\", \"opusplan\", \"haiku\", \"gpt-5.4\", \"gpt-5.5\", \"gpt-5.4-mini\", \"gpt-5.3-codex\", \"gpt-5.3-codex-spark\", \"gpt-5.2\", \"gemini-2.5-pro\", \"gemini-2.5-flash\", \"gemini-3.1-pro-preview\", \"gemini-3-pro-preview\", \"gemini-3-flash-preview\", \"forge\", \"opencode\". OpenCode also accepts explicit dynamic models using \"oc-<provider/model>\". \"forge\" is a provider key, not a Forge model family selector."New value: +"The model to use. Aliases: \"claude-ultra\" (Opus with auto max effort; does not select Fable), \"codex-ultra\" (auto ultra reasoning), \"gemini-ultra\". Standard: \"sonnet\", \"sonnet[1m]\", \"opus\", \"opusplan\", \"fable\", \"haiku\", \"gpt-5.4\", \"gpt-5.6-sol\", \"gpt-5.6-terra\", \"gpt-5.6-luna\", \"gpt-5.5\", \"gpt-5.4-mini\", \"gpt-5.3-codex\", \"gpt-5.3-codex-spark\", \"gpt-5.2\", \"gemini-2.5-pro\", \"gemini-2.5-flash\", \"gemini-3.1-pro-preview\", \"gemini-3-pro-preview\", \"gemini-3-flash-preview\", \"forge\", \"opencode\". Fable is an explicit selection and may require usage credits. OpenCode also accepts explicit dynamic models using \"oc-<provider/model>\". \"forge\" is a provider key, not a Forge model family selector."
    • changedInput schema / properties / reasoning_effort / description
      Previous value: -"Reasoning control for Claude and Codex. Claude uses --effort with \"low\", \"medium\", \"high\", \"xhigh\", \"max\". Codex uses model_reasoning_effort with \"low\", \"medium\", \"high\", \"xhigh\". Gemini, Forge, and OpenCode do not support reasoning_effort in this integration."New value: +"Reasoning control for Claude and Codex. Claude uses --effort with \"low\", \"medium\", \"high\", \"xhigh\", \"max\". Codex uses model_reasoning_effort with \"low\", \"medium\", \"high\", \"xhigh\"; GPT-5.6 Sol and Terra also support \"max\" and \"ultra\", while Luna supports \"max\". Gemini, Forge, and OpenCode do not support reasoning_effort in this integration."
  7. First observedv2.21.0

TDQS

A4/5.0
Behavior4/5

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

With no annotations, the description carries the full burden. It discloses the key async trait ('returns immediately with a PID', 'runs in the background'), instructs the agent to use other tools for status/results, and mentions killing long-running processes. It doesn't warn about side effects of arbitrary terminal commands or process lifetime limits, but these are minor for an agent-runner.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness3/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The purpose, capabilities, and tips are front-loaded and well-structured. However, the 'Supported models' section repeats what the model parameter schema already documents in detail, adding a large block of redundancy. The capability bullet list is useful but generic, so not every sentence earns its place.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a complex async runner with six parameters and no output schema, the definition is largely complete: it explains background execution, immediate PID return, the monitoring workflow, prompt constraints, and actionable tips. It omits details like process cleanup policy or error reporting, but the sibling tools (get_result, kill_process, list_processes) are referenced to cover those workflows.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100% with detailed descriptions for all six parameters, so the baseline is 3. The description adds the prompt/prompt_file exclusivity rule and general prompt tips, but the long supported-models list largely duplicates the schema's model parameter description. Added semantic value beyond the schema is minimal.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description opens with a specific verb and resource: 'Starts a Claude, Codex, Antigravity (Gemini), OpenCode, Grok, or Pi CLI process in the background and returns a PID immediately.' It also differentiates from siblings by directing to list_processes and get_result for monitoring, and it lists concrete capability domains (file ops, code, git, terminal, web).

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

It gives explicit workflow guidance: monitor progress with list_processes, retrieve results with get_result, and kill with kill_process; these are also restated in numbered prompt tips. It clearly states the prompt OR prompt_file constraint. However, it doesn't explicitly state when not to use this tool or name an alternative runner, so it stops short of full when/when-not coverage.

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

Deploy Server

Other Tools