Gemini
Route one-shot prompts, JSON outputs, extensions, skills, hooks, or MCP through the Gemini CLI from your existing agent workflow.
Overview
Gemini is an agent skill most often used in Build (also Idea research and Grow content) that documents headless Gemini CLI prompts, JSON output, and extension/MCP commands.
Install
npx skills add https://github.com/steipete/clawdis --skill geminiWhat is this skill?
- Headless one-shot mode via `-p/--prompt`; positional args open interactive session
- Model flag `-m`, JSON output via `--output-format json`, stdin pipes into the prompt
- Subcommands for extensions, skills, hooks, and MCP management
- Documents `gemini --list-extensions` and install path via Homebrew `gemini-cli`
- Explicitly avoids `--yolo` for safer defaults
Adoption & trust: 1.8k installs on skills.sh; 378k GitHub stars; 3/3 security scanners passed (skills.sh audits).
What problem does it solve?
You want Gemini or Gemma in a scriptable one-shot CLI form but keep opening interactive sessions or missing JSON-friendly flags.
Who is it for?
Indie builders already using terminal agents who need quick Gemini passes, JSON extraction, or CLI extension management alongside Claude or Codex.
Skip if: Teams without the `gemini` binary installed or workflows that require only in-IDE chat with no shell access.
When should I use this skill?
User wants Gemini CLI one-shot prompts, summaries, generation, skills, hooks, MCP, or Gemma routing from the terminal.
What do I get? / Deliverables
You run documented `gemini -p` (and related) commands with the right model, output format, and extension/MCP hooks after one-time CLI auth.
- Terminal command recipes for `-p` one-shot and JSON output
- Extension/skills/hooks/MCP CLI command references
- Model-selected prompt results from stdin or inline text
Recommended Skills
Journey fit
Spans multiple journey phases - primary shelf plus alternate fits below.
Gemini CLI is installed and invoked while assembling the agent toolchain alongside other model CLIs. The skill documents headless Gemini invocation and CLI subcommands—not app frontend or infra monitoring.
Where it fits
Summarize competitor notes piped from a markdown file with `gemini -p` before scoping the build.
Return structured JSON from Gemini for a small codegen or config task in a headless CI script.
Draft or refine marketing copy with a chosen `-m` model in one-shot mode.
How it compares
CLI routing cheat sheet for Gemini—not a hosted MCP server or full multi-agent planner by itself.
Common Questions / FAQ
Who is gemini for?
Solo builders and agent users who have or can install Gemini CLI and want reliable one-shot and JSON invocation from automation.
When should I use gemini?
In Idea for quick research summaries; in Build for generation or JSON tasks via `-p`; when configuring extensions, skills, hooks, or MCP; whenever stdin plus prompt batching is needed.
Is gemini safe to install?
It shells out to Google’s CLI; review the Security Audits panel on this Prism page and avoid unsafe flags such as `--yolo` as the skill recommends.
SKILL.md
READMESKILL.md - Gemini
# 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 `gemini` once interactively and follow the login flow. - Avoid `--yolo` for safety.