skills.rest
Kimjiman
KimjimanCOMMUNITY
Explore all skills by this author →
GitHub

iterative-retrieval

Iteratively retrieve and evaluate codebase files to refine agent context.

Updated May 24, 2023
Search Tags:#context#codebase#RAG#agents#retrieval#iterative
One-click install
npx skills add https://github.com/Kimjiman/basic-arch --skill iterative-retrieval-kimjiman
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: iterative-retrieval
Source: https://github.com/Kimjiman/basic-arch/tree/main/.claude/skills/iterative-retrieval
Command: npx skills add https://github.com/Kimjiman/basic-arch --skill iterative-retrieval-kimjiman

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill addresses the challenge where subagents need codebase context that they cannot predict upfront, by progressively refining retrieval to solve the subagent context problem.

Core Features & Use Cases

  • Progressive Context Refinement: Iteratively searches and evaluates codebase files to gather the most relevant context for agent tasks.
  • Handles Unpredictable Context Needs: Ideal for scenarios where subagents require specific code snippets or file paths that aren't known at the start of the workflow.
  • Use Case: When debugging a complex issue in a large codebase, this Skill can help an agent discover the exact files and code sections related to the bug without needing an exhaustive initial search.

Quick Start

Use the iterative-retrieval skill to find files related to user authentication and session management.

Frequently Asked Questions about iterative-retrieval

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

FAQPage Schema
How do I dynamically retrieve codebase context for agent tasks when needed files are unknown?▼

Iterative retrieval progressively searches and evaluates codebase files to gather relevant context for agent tasks. It dynamically identifies necessary file paths based on evolving task requirements, solving unpredictable subagent context needs.

What is the best way to refine context for complex agent workflows in a large codebase?▼

Refining context for complex agent workflows uses a multi-phase iterative retrieval pattern. This approach dynamically identifies and gathers relevant codebase files, optimizing token usage through targeted information gathering instead of exhaustive initial searches.

Can I use iterative retrieval to build dynamic RAG-like pipelines for software engineering?▼

Yes, iterative retrieval supports scenarios requiring dynamic RAG-like pipelines. It progressively refines retrieval to solve the subagent context problem, allowing agents to discover exact code sections related to specific issues without upfront knowledge.

How do I optimize token usage when subagents need to search a large codebase?▼

Optimize token usage by implementing targeted information gathering through iterative retrieval. This pattern progressively refines context by dynamically identifying and gathering only the relevant codebase files required for the evolving agent task.

When do I need progressive context refinement for debugging complex issues?▼

Progressive context refinement is needed when debugging complex issues in a large codebase where specific file paths and code snippets are unpredictable. Iterative retrieval helps agents discover exact related files dynamically without exhaustive initial search.

Related Skills

Explore More Software Engineering Skills→
tneztnez

bootstrap

Initialize the .docent directory structure and configuration for new projects.

Community
Basic
alanmunalanmun

debug

Analyze code errors and generate patches from error logs.

Community
Intermediate
artsmcartsmc

start-phase-execute-team

Coordinate parallel task execution for multi-agent teams across start-phase workflows.

Community
Advanced
cyphercr0wcyphercr0w
8

iterative-retrieval

Iteratively search and evaluate codebase files to build relevant context sets.

Community
Advanced