generic-subagent

Delegate read-only research to subagents and return structured digests.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/Abkr1/liza --skill generic-subagent
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: generic-subagent
Source: https://github.com/Abkr1/liza/tree/main/skills/generic-subagent
Command: npx skills add https://github.com/Abkr1/liza --skill generic-subagent

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill enables context-efficient delegation by offloading read-only research to subagents, returning a concise digest instead of full traces and ensuring state is preserved.

Core Features & Use Cases

  • Read-only delegation: Offloads large, context-heavy analysis to subagents without mutating state.
  • Digest output: Returns a structured digest suitable for integration with the main task.
  • Triggers and thresholds: Applies rules such as when content to read exceeds 250KB and processing depth > 2 steps.
  • Use Case: Ideal for codebase reviews, large-document analyses, and knowledge extraction tasks where full context is expensive to compute.

Quick Start

Provide a MODE: SUBAGENT brief with a clear objective, e.g., MODE: SUBAGENT; GOAL: summarize the dependencies of project/*.py; CONTEXT: minimal prior knowledge.

Frequently Asked Questions about generic-subagent

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

FAQPage Schema
How do I delegate read-only code analysis tasks without consuming too much context?

Read-only code analysis delegation offloads large context-heavy research to subagents, returning a concise structured digest instead of full traces to preserve context. This ensures state is not mutated while extracting knowledge from large codebases efficiently.

When should I use subagent delegation for large document summarization?

Subagent delegation for large document summarization is needed when content exceeds 250KB and processing depth requires more than two steps. It applies to scenarios like documentation reviews and knowledge extraction where full context computation is expensive.

What is the best way to summarize project dependencies across multiple files?

The best way to summarize project dependencies is providing a MODE: SUBAGENT brief with a clear objective and minimal prior knowledge context. The subagent enforces a read-only constraint and outputs a structured digest suitable for downstream tasks.

Does subagent delegation modify the codebase state during code audits?

Subagent delegation does not modify codebase state during code audits because it enforces a strict read-only constraint. It follows the Subagent Delegation Protocol to ensure state is preserved while returning a structured digest for review.

What are the limitations of using read-only subagent delegation for research tasks?

Read-only subagent delegation is limited to research and summarization tasks because it enforces a read-only constraint and cannot mutate state. It is not suitable for tasks requiring state modification or direct code execution within the main context.