skills.rest
TruCol270
TruCol270COMMUNITY
Explore all skills by this author →
GitHub

iterative-retrieval

Iteratively refine codebase context for subagents through a four-phase retrieval loop.

1|Updated Apr 7, 2026
Search Tags:#workflow#context#codebase#subagents#multi-agent#retrieval#iterative
One-click install
npx skills add https://github.com/TruCol270/salty-pickle --skill iterative-retrieval-trucol270
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: iterative-retrieval
Source: https://github.com/TruCol270/salty-pickle/tree/main/.claude-skills/iterative-retrieval
Command: npx skills add https://github.com/TruCol270/salty-pickle --skill iterative-retrieval-trucol270

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Subagents often start with limited or missing context, which leads to inefficient task execution or incorrect results. This skill provides a structured approach to progressively refine and accumulate relevant context from the codebase to enable effective subagent operation.

Core Features & Use Cases

  • Four-phase iterative retrieval loop (DISPATCH, EVALUATE, REFINE, LOOP) to progressively narrow context.
  • Context-aware file selection with relevance scoring and cycle-based refinement to handle large or evolving codebases.
  • Use cases include autonomous code exploration, multi-agent task orchestration, and workflows with strict token/context limits.

Quick Start

Start the iterative retrieval loop to progressively refine codebase context for subagents.

Frequently Asked Questions about iterative-retrieval

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

FAQPage Schema
How do I refine codebase context for subagents before task execution?▼

You can progressively refine codebase context for subagents by running a four-phase iterative retrieval loop that evaluates and narrows relevant files before execution begins.

What is the best way to handle context limits when orchestrating multi-agent workflows?▼

Iterative retrieval handles context limits by applying context-aware file selection with relevance scoring, progressively narrowing down large codebases across a maximum of three cycles.

How does the iterative retrieval loop resolve missing context in multi-agent systems?▼

The iterative retrieval loop resolves missing context through a DISPATCH, EVALUATE, REFINE, and LOOP sequence that progressively accumulates relevant codebase information using explicit relevance-driven refinement and gating rules.

Can I use this approach for autonomous code exploration in evolving codebases?▼

Yes, this approach supports autonomous code exploration and multi-agent task orchestration by using cycle-based refinement to adapt context retrieval when working with large or evolving codebases.

What are the limitations of using a loop-based context refinement pattern for subagents?▼

The loop-based context refinement pattern is limited to a maximum of three cycles, meaning subagents must achieve sufficient context relevance within that constrained iteration boundary.

Related Skills

Explore More Software Engineering Skills→
busboom-software-studiobusboom-software-studio
1

dispatch-subagent

Dispatch subagents with curated context and strict directory scope.

Official
Intermediate
klu-devklu-dev
2

iterative-retrieval

Refine context progressively across four phases for multi-agent workflows.

Community
Advanced
vinitgirdharvinitgirdhar

iterative-retrieval

Refine subagent context through a four-phase retrieval loop.

Community
Intermediate
bennybennisonbennybennison

iterative-retrieval

Run progressive multi-cycle codebase searches with relevance scoring and terminology refinement.

Community
Intermediate