skills.rest
RambleRainbow
RambleRainbowCOMMUNITY
Explore all skills by this author →
GitHub

iterative-retrieval

Retrieve codebase context through a 4-phase iterative loop with relevance scoring.

Updated May 9, 2026
Search Tags:#agent orchestration#code exploration#context optimization#rag pipeline#iterative retrieval#codebase search#subagent context
One-click install
npx skills add https://github.com/RambleRainbow/jd --skill iterative-retrieval-ramblerainbow
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: iterative-retrieval
Source: https://github.com/RambleRainbow/jd/tree/main/.claude/skills/iterative-retrieval
Command: npx skills add https://github.com/RambleRainbow/jd --skill iterative-retrieval-ramblerainbow

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill resolves the common failure mode in multi-agent workflows where subagents lack sufficient codebase context to complete tasks, avoiding the tradeoff between sending excessive context that exceeds token limits and sending insufficient context that leads to missing information and failed executions.

Core Features & Use Cases

  • 4-Phase Iterative Loop: Implements a structured Dispatch, Evaluate, Refine, and Loop workflow to progressively narrow down relevant context.
  • Relevance Scoring: Evaluates retrieved files against a 0-1 scale to prioritize high-value context and exclude irrelevant content.
  • Use Cases: Ideal for bug fix context gathering, feature implementation research, RAG-like code exploration pipelines, and optimizing token usage in agent orchestration.

Quick Start

Use the iterative-retrieval skill to gather the precise codebase context your subagent needs for its task without overwhelming its context window or missing critical information.

Frequently Asked Questions about iterative-retrieval

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

FAQPage Schema
How do I fix subagent context gaps in multi-agent code workflows?▼

Subagent context gaps occur when agents lack sufficient codebase information to complete tasks. A progressive 4-phase retrieval loop solves this by dispatching, evaluating, refining, and looping to gather precise context without exceeding token limits.

What's the best way to optimize token usage for agent orchestration during codebase search?▼

Token optimization during codebase search requires progressive retrieval with relevance scoring. By evaluating retrieved files against a 0-1 scale, you prioritize high-value context and exclude irrelevant content, preventing subagent context window exhaustion.

How does progressive retrieval work for RAG pipeline code exploration?▼

Progressive retrieval for RAG code exploration uses a 4-phase iterative loop. It dispatches search criteria, evaluates context relevance, refines missing context gaps, and loops dynamically to narrow down relevant codebase files for subagent execution.

Does iterative retrieval cap resource usage when gathering bug fix context?▼

Yes, iterative retrieval caps resource usage with a maximum 3-cycle iteration limit. This prevents excessive search loops while ensuring enough codebase context is gathered for subagent bug resolution and feature development tasks.

Why does my subagent fail tasks due to insufficient codebase context?▼

Subagents fail when sent insufficient codebase context, missing critical information for execution. Dynamic search criteria refinement identifies missing context gaps and progressively retrieves relevant files to ensure complete task context.

Related Skills

Explore More Software Engineering Skills→
xu-xiangxu-xiang
1.8k

iterative-retrieval

Refine repository context through iterative retrieval cycles for subagent coordination.

Community
Basic
sumeetonline90sumeetonline90

iterative-retrieval

Refine retrieval queries across up to 3 cycles to identify relevant codebase files.

Community
Intermediate
sayasaya8039sayasaya8039
2

iterative-retrieval

Retrieve relevant codebase context for subagents through iterative refinement cycles.

Community
Intermediate
rafazshrafazsh

iterative-retrieval

Iteratively search and refine relevant files for software engineering tasks.

Community
Intermediate