Gemini
- 3k installs
- 385k repo stars
- Updated August 3, 2026
- steipete/clawdis
gemini is an OpenClaw skill for headless Gemini CLI prompts, JSON output, and extensions, skills, hooks, and MCP subcommands.
About
The gemini skill wraps Google's Gemini CLI in headless one-shot mode for agents using the gemini binary installed via brew formula gemini-cli on macOS OpenClaw hosts. Positional arguments start interactive mode, so agents should use -p or --prompt for non-interactive runs, optionally with -m model selection and --output-format json for structured responses. Stdin can append to a prompt, enabling piped summaries from files such as notes.md through shell pipes. Extension management commands list, install, and manage gemini extensions, skills, hooks, and MCP integrations through documented subcommands. Notes require one-time interactive auth if needed and explicitly warn against --yolo for safety. OpenClaw metadata specifies darwin OS, required gemini bin, and brew install instructions with emoji and homepage links. Typical uses include quick question answering, JSON extraction, extension discovery, and routing to Gemini skills or MCP tooling without launching a full interactive terminal session.
- Headless gemini -p one-shot prompts instead of interactive positional mode.
- JSON output format flag for structured agent parsing.
- Stdin piping to append file content into prompts.
- Extensions, skills, hooks, and MCP subcommand routing.
- Brew install metadata and safety note against --yolo flag.
Gemini by the numbers
- 3,002 all-time installs (skills.sh)
- +157 installs in the week ending Aug 2, 2026 (Skillselion tracking)
- Ranked #257 of 16,556 AI & Agent Building skills by installs in the Skillselion catalog
- Security screen: LOW risk (skills.sh audit)
- Data as of Aug 3, 2026 (Skillselion catalog sync)
gemini capabilities & compatibility
- Capabilities
- headless p prompt execution · model flag and json output formatting · stdin append piping for file summaries · extension, skill, hook, and mcp subcommands · brew install and binary requirement metadata · safety guidance against yolo flag
- Works with
- openai
- Use cases
- orchestration · translation · research
- Platforms
- macOS
- Runs
- Runs locally
- Pricing
- Bring your own API key
What gemini says it does
Positional text starts interactive mode; use `-p/--prompt`.
`gemini -p "Return JSON" --output-format json`
Avoid `--yolo` for safety.
npx skills add https://github.com/steipete/clawdis --skill geminiAdd your badge
Show developers this skill is listed on Skillselion. Paste this into your README.
| Installs | 3k |
|---|---|
| repo stars | ★ 385k |
| Security audit | 3 / 3 scanners passed |
| Last updated | August 3, 2026 |
| Repository | steipete/clawdis ↗ |
How do agents run Gemini CLI non-interactively with optional JSON output and extension tooling?
Run Gemini CLI one-shot prompts, JSON output, extensions, skills, hooks, and MCP commands from agents.
Who is it for?
macOS agent setups with gemini-cli installed that need quick Gemini calls or CLI extension management.
Skip if: Skip for direct Google AI Studio API integration without the gemini binary, or non-macOS hosts without the CLI.
When should I use this skill?
User wants Gemini one-shot prompts, summaries, JSON generation, or gemini extensions, skills, hooks, or MCP commands.
What you get
Gemini CLI response text or JSON from a one-shot -p prompt, with optional extension or MCP command routing.
- Headless Gemini responses
- CLI one-shot outputs
- MCP-routed Gemini calls
By the numbers
- Requires gemini binary installed via Homebrew formula gemini-cli
Files
Gemini CLI
Use Gemini in headless one-shot mode. Positional text starts interactive mode; use -p/--prompt.
Quick start
gemini -p "Answer this question..."gemini -m <model> -p "Prompt..."gemini -p "Return JSON" --output-format json- stdin appends to
-p:cat notes.md | gemini -p "Summarize"
Extensions
- List:
gemini --list-extensions - Manage:
gemini extensions <command> - Skills:
gemini skills <command> - Hooks:
gemini hooks <command> - MCP:
gemini mcp <command>
Notes
- If auth is required, run
geminionce interactively and follow the login flow. - Avoid
--yolofor safety.
Related skills
Setup Matt Pocock SkillsScaffold the per-repo configuration that Matt Pocock’s engineering agent skills rely on so they understand the issue tracker, triage labels, and domain documentation la495k196k
Lark AppsConnect Claude, Cursor or custom agents directly to Lark (Feishu) for messaging, document automation, approval workflows and enterprise data access.402k
CavemanSlash token usage by roughly 75% while keeping every technical detail intact when working with Claude Code, Cursor or similar agents.391k94.7k
Lark Skill MakerQuickly turn any Lark/Feishu OpenAPI call or multi-step workflow into a reusable agent skill with its own SKILL.md.386k16k
Running Claude Code Via Litellm CopilotRun Claude Code at a fraction of the cost by routing requests through LiteLLM to the GitHub Copilot Chat API.270k72
Codex PetGenerate a complete Codex Pet spritesheet and metadata from one reference image without needing an OpenAI key or Codex Pro.246k10How it compares
Pick gemini over embedded SDK calls when you want a brew-installed CLI with repeatable gemini -p one-shot commands inside shell-driven agent workflows.
FAQ
How do I avoid interactive Gemini mode?
Use -p or --prompt with your text; positional-only invocation starts interactive mode.
How do I get JSON from Gemini CLI?
Add --output-format json to a gemini -p prompt invocation.
How is gemini CLI installed?
Via brew formula gemini-cli per OpenClaw metadata, producing the gemini binary.
Is Gemini safe to install?
skills.sh reports 3 of 3 security scanners passed. Review the Security Audits panel on this page before installing in production.