skills.rest
itou-daiki
itou-daikiCOMMUNITY
Explore all skills by this author →
GitHub

iterative-retrieval

Retrieve relevant files through iterative search, evaluation, and refinement cycles.

Updated Nov 17, 2025
Search Tags:#workflow#subagents#context-management#information-retrieval#iterative-retrieval#ai-augmentation
One-click install
npx skills add https://github.com/itou-daiki/easy_stat_edu --skill iterative-retrieval-itou-daiki
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: iterative-retrieval
Source: https://github.com/itou-daiki/easy_stat_edu/tree/main/.agent/skills/iterative-retrieval
Command: npx skills add https://github.com/itou-daiki/easy_stat_edu --skill iterative-retrieval-itou-daiki

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Refines context for subagents by iteratively searching, evaluating, and refining files to assemble relevant information with limited upfront context.

Core Features & Use Cases

  • 4-phase loop (DISPATCH, EVALUATE, REFINE, LOOP) to progressively improve context with a capped number of cycles.
  • Relevance scoring and gap reporting to focus only on high-value files and identify missing information.
  • Safe, deterministic workflow suitable for integration into multi-agent tasks and debugging.
  • Use Case: When a subagent starts with incomplete context, apply iterative retrieval to surface the most relevant code, docs, and data.

Quick Start

Provide a short task description and execute the iterative retrieval process to obtain a compact set of highly relevant files within three cycles.

Frequently Asked Questions about iterative-retrieval

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

FAQPage Schema
How do I retrieve relevant context for subagents working across multi-file codebases?▼

Iterative retrieval identifies relevant files by progressively searching, evaluating, and refining context. It assembles necessary information for subagents even when starting with limited upfront context across multi-file codebases.

Can I control how many search cycles a subagent runs to gather context?▼

Yes, the iterative retrieval workflow enforces a maximum of three cycles. This cap ensures safe, predictable context augmentation by limiting the iterative search, evaluation, and refinement loop.

What is the best way to find missing information when a subagent has incomplete context?▼

Use iterative retrieval to apply explicit gap reporting. This mechanism evaluates retrieved files, identifies missing information, and focuses subsequent search rounds only on high-value files to fill the gaps.

Does iterative retrieval work for debugging tasks in complex codebases?▼

Yes, iterative retrieval is suitable for debugging and multi-agent tasks. It provides a safe, deterministic workflow to progressively surface relevant code, docs, and data when a subagent starts with incomplete context.

Why does my subagent fail to solve problems with incomplete context in large codebases?▼

Subagents fail when they lack sufficient relevant files to understand the problem space. Iterative retrieval solves this by progressively searching and evaluating files, using relevance scoring to focus only on high-value information.

Related Skills

Explore More Software Engineering Skills→
VenTheZoneVenTheZone
1

iterative-retrieval

Iteratively search and evaluate codebase files to refine context retrieval.

Community
Advanced
himanshu231204himanshu231204
2

iterative-retrieval

Refine codebase context retrieval through a 4-phase dispatch-evaluate-refine loop.

Community
Intermediate
frankxaifrankxai
6

orchestration/parallel-execution

Coordinate parallel task execution across agents and tools with thread-safe dispatch and failure recovery.

Community
Intermediate
rafazshrafazsh

iterative-retrieval

Iteratively search and refine relevant files for software engineering tasks.

Community
Intermediate