gemini-delegator

Name: gemini-delegator
Availability: InStock
Author: nexus-substrate

Delegate large-context and multimodal analysis tasks to the Gemini CLI.

18|1|Updated Jan 3, 2026

Search Tags:#routing #multimodal #codebase analysis #gemini #json output #context window #delegate

Ready to run

npx skills add https://github.com/nexus-substrate/nexus-agents --skill gemini-delegator-nexus-substrate

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: gemini-delegator
Source: https://github.com/nexus-substrate/nexus-agents/tree/main/skills/gemini-delegator
Command: npx skills add https://github.com/nexus-substrate/nexus-agents --skill gemini-delegator-nexus-substrate

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It solves the challenge of analyzing extremely large contexts and multimodal inputs without timing out or overloading your primary model.

Core Features & Use Cases

Large-context delegation: Offloads tasks when input context exceeds about 100K tokens, enabling whole-codebase or multi-document analysis that would be impractical elsewhere.
Multimodal routing: Handles image and screenshot understanding, including extracting UI details or text from visual content.
Flexible execution modes: Supports intelligent routing via an MCP tool or direct Gemini CLI calls, including JSON outputs for parsing pipelines.

Quick Start

Delegate this task to Gemini by saying: delegate to gemini to analyze my 500K token codebase and return a concise structured summary.

Frequently Asked Questions about gemini-delegator

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

FAQPage Schema

How do I analyze a large codebase that exceeds 100K tokens without timing out?▼

Whole-codebase analysis for inputs over 100K tokens is handled by routing the task to the Gemini CLI, which processes large contexts faster and returns concise structured summaries without timing out.

How does multimodal routing work for image and screenshot analysis?▼

Multimodal routing detects visual inputs like images and screenshots, then delegates the analysis to Gemini to extract UI details or text from visual content efficiently.

Can I get JSON output from delegated tasks for parsing pipelines?▼

Yes, delegated tasks support JSON-friendly output handling, allowing the Gemini CLI to return structured data directly for seamless integration into automated parsing pipelines.

Does this approach support both direct CLI calls and MCP tool routing?▼

Flexible execution modes support intelligent routing via an MCP tool or direct Gemini CLI calls, allowing you to choose the best method for your specific batch processing workflow.

When should I delegate batch processing tasks to Gemini instead of using my primary model?▼

Delegate to Gemini when tasks are speed or cost-sensitive, require multimodal understanding, or involve context sizes exceeding 100K tokens that would overload your primary model.