gemini-agent
Skill8 starsChanged 8 months agoMITView on GitHub
Delegates a task to Google Gemini CLI as a subagent and returns its response. Use this skill whenever the user wants Gemini's opinion, a second opinion from Gemini, wants to ask Gemini to do something, wants to compare Claude vs Gemini on a task, or says anything like "ask Gemini", "have Gemini look at this", "what does Gemini think", "run this through Gemini", or "get Gemini to review". Also use when the user wants multiple AI perspectives on code, writing, research, or analysis — Gemini has a very large context window useful for long documents.
- Reads credentials
- Installs packages
2 other projects use a version of this.
Adopted by
Most recently taken up by
- ZacheryGlass/.claudetheir own version, first seen Nov 1, 2026
- ThendCN/ai-team-skillstheir own version, first seen Oct 16, 2026
Other versions2 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 2
- ZacheryGlass/.claude★ 599same name
- ThendCN/ai-team-skills★ 96same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository5 other files it gives its agents.
- AGENTS.mdAGENTS.md
- agentic-radarSkill
- codex-agentSkill
- reflect-and-learnSkill
- vendor-docs-radarSkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 0 are distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 24
Always pass -m gemini-3.1-pro-preview-customtools unless speed matters more than quality, in which case use -m …
- Very specific: line 32
Use -p / --prompt for non-interactive (headless) mode:
- Very specific: line 60
Run via Bash tool: gemini -m gemini-3.1-pro-preview-customtools -p "<prompt>" or cat file | gemini -m gemini-3.1-pro-preview-customtools -p …
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Gemini Agent Skill
Runs a prompt through the Gemini CLI non-interactively and returns the full response.
Models
| Purpose | Model | When to use |
| Default (agentic work) | gemini-3.1-pro-preview-customtools | Multi-step tasks, tool use, code generation, reviews, research |
| Fallback / fast | gemini-3.1-flash-lite-preview | Simple questions, quick lookups, low-latency needs |
Always pass -m gemini-3.1-pro-preview-customtools unless speed matters more than quality,
in which case use -m gemini-3.1-flash-lite-preview.
Note: Model names change frequently. Update the table above when new models are released. Check
gemini --list-modelsor the vendor-docs-radar reports for the latest.
How to invoke
Use -p / --prompt for non-interactive (headless) mode:
gemini -m gemini-3.1-pro-preview-customtools -p "<prompt>"Or pipe content via stdin with the prompt appended:
cat <file> | gemini -m gemini-3.1-pro-preview-customtools -p "<instruction>"Key flags
-p "<prompt>"— run non-interactively, prints response to stdout-m <model>— override model-y/--yolo— auto-approve all actions (useful for non-interactive batch tasks)--approval-mode yolo— same as -y, auto-approve everything
Thinking level
For complex tasks (research synthesis, architecture review, multi-step reasoning),
pass thinking_level: high if the CLI or API supports it. This triggers extended
chain-of-thought and improves accuracy on hard problems.
Workflow
- Formulate the prompt — include all context Gemini needs. Gemini has a very large context window (1M+ tokens), so you can pipe entire long documents via stdin.
- Run via Bash tool:
gemini -m gemini-3.1-pro-preview-customtools -p "<prompt>"orcat file | gemini -m gemini-3.1-pro-preview-customtools -p "<instruction>" - Capture stdout as the response
- Return Gemini's full response to the user, clearly attributed as "Gemini says:"
Passing file content
For short content, inline it:
gemini -m gemini-3.1-pro-preview-customtools -p "Review this paper abstract:\n\n<content here>"For long documents (papers, codebases), pipe via stdin — Gemini excels at long context:
cat /path/to/paper.tex | gemini -m gemini-3.1-pro-preview-customtools -p "Give a critical review of this paper. Focus on: methodology, claims, and weaknesses."Notes
- Gemini runs non-interactively with
-p; responses stream to stdout - For very long documents, Gemini's large context window is a key advantage over other CLIs
- The response comes from Google's Gemini models; clearly label it as coming from Gemini when presenting to the user
- May take 30-90 seconds for complex or long-context tasks
- Deprecated models (do not use):
gemini-2.0-flash,gemini-2.5-pro,gemini-3-pro-preview— these are shut down or end-of-life
Prerequisites
- Install Gemini CLI:
npm install -g @google/gemini-clior follow https://github.com/google-gemini/gemini-cli - Authenticate: run
geminionce interactively to complete OAuth or setGEMINI_API_KEY
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

