iterative-retrieval

Iteratively discover and evaluate code files to refine context for subagents.

Updated Dec 4, 2025

Search Tags:#context #codebase #subagents #RAG #retrieval #iterative

Ready to run

npx skills add https://github.com/hiroto0701/dotfiles --skill iterative-retrieval-hiroto0701

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: iterative-retrieval
Source: https://github.com/hiroto0701/dotfiles/tree/main/private_dot_claude/skills/iterative-retrieval
Command: npx skills add https://github.com/hiroto0701/dotfiles --skill iterative-retrieval-hiroto0701

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill addresses the challenge where subagents require codebase context that they cannot predict upfront, by progressively refining the retrieved information.

Core Features & Use Cases

Progressive Context Refinement: Iteratively searches and evaluates code files to build relevant context.
Handles Unpredictable Needs: Ideal for scenarios where subagents need to discover necessary code patterns or terminology as they work.
Use Case: When debugging a complex feature, this Skill can help an agent find all relevant authentication and session management files even if it doesn't know their exact names or locations initially.

Quick Start

Use the iterative-retrieval skill to find all files related to user authentication and session management within the project's source directory.

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 subagents when file locations are unknown?▼

Dynamically retrieving codebase context for subagents is handled through a multi-phase iterative retrieval pattern that progressively searches and evaluates files to build relevant context without upfront knowledge of file names or locations.

What is iterative retrieval in the context of agent orchestration and token optimization?▼

Iterative retrieval in agent orchestration is a dynamic RAG-like pipeline that progressively refines context for subagents, optimizing token usage by discovering only the codebase elements relevant to the current workflow phase.

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

Progressive context refinement is needed when debugging complex features where subagents cannot predict upfront which files are relevant, such as discovering all authentication and session management files scattered across a source directory.

Can I use iterative retrieval with existing RAG pipelines for code exploration?▼

Iterative retrieval can support dynamic RAG-like pipelines for code exploration by iteratively searching and evaluating code files, allowing subagents to discover necessary code patterns and terminology as they work through complex workflows.

What are the limitations of iterative retrieval for unpredictable subagent context needs?▼

Iterative retrieval requires multiple search phases to evaluate and refine context, which may increase latency compared to direct file access, but is necessary when subagents lack upfront knowledge of required codebase elements.