cli-gemini

Delegate tasks to Google's Gemini CLI for web search and codebase analysis.

31|3|Updated Dec 23, 2025
One-click install
npx skills add https://github.com/MichelKerkmeester/opencode--spec-kit-skilled-agent-orchestration --skill cli-gemini
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: cli-gemini
Source: https://github.com/MichelKerkmeester/opencode--spec-kit-skilled-agent-orchestration/tree/main/.opencode/skill/cli-gemini
Command: npx skills add https://github.com/MichelKerkmeester/opencode--spec-kit-skilled-agent-orchestration --skill cli-gemini

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes assets (resource) and references (resource) components.

What problem does it solve?

Gemini CLI orchestration enables any AI to delegate specialized tasks to Gemini for live web research, codebase analysis, and multi-agent workflows inside a unified terminal session, reducing internal tool friction and boosting speed.

Core Features & Use Cases

  • Cross-AI delegation: route tasks to Gemini agents (e.g., @context, @review, @deep-research, @write, @ultra-think, @debug) for targeted work.
  • Live information grounding: leverage google_web_search to fetch current sources and citations.
  • Architecture and codebase analysis: map dependencies and generate architecture docs with codebase_investigator.
  • Memory and continuity: save_memory for cross-session context and Spec Kit memory integration.
  • Parallel task processing: offload background work while continuing other tasks.

Quick Start

Install the Gemini CLI and invoke a sample prompt using the @<agent> routing to observe read-only exploration and structured outputs.

Frequently Asked Questions about cli-gemini

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How does cross-AI delegation work with the Gemini CLI?

Cross-AI delegation routes specialized tasks to Gemini CLI agents like @deep-research or @debug. It enables any AI to offload work, such as live web research or codebase analysis, through a unified terminal session.

How do I generate architecture docs and map dependencies using a CLI?

To generate architecture docs and map dependencies, you delegate tasks to the codebase_investigator agent. This agent analyzes your codebase structure and outputs architectural mappings directly within your terminal session.

Can I use live web search grounding for AI tasks inside my terminal?

Yes, live web search grounding is supported via google_web_search. It fetches current sources and citations directly within the terminal session to ground delegated AI tasks in real-time web data.

Does cross-AI orchestration prevent circular delegation loops?

Yes, cross-AI orchestration enforces a self-invocation guard. This mechanism prevents circular delegation loops, ensuring safe and continuous parallel task processing without recursive agent calls.

What is the best way to maintain memory continuity across multiple AI sessions?

The best way to maintain memory continuity is using the save_memory feature. It preserves cross-session context and integrates with Spec Kit memory to retain state between different terminal workflows.