iterative-retrieval

Iteratively refine codebase context retrieval across up to three search loops.

3|Updated Jul 28, 2025
One-click install
npx skills add https://github.com/dileep-kishore/zendots --skill iterative-retrieval-dileep-kishore
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: iterative-retrieval
Source: https://github.com/dileep-kishore/zendots/tree/main/private_dot_claude/skills/iterative-retrieval
Command: npx skills add https://github.com/dileep-kishore/zendots --skill iterative-retrieval-dileep-kishore

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill addresses the challenge where subagents require codebase context that they cannot predict upfront, preventing them from starting tasks effectively.

Core Features & Use Cases

  • Progressive Context Refinement: Iteratively searches and evaluates files to build the most relevant context for agent tasks.
  • Handles Unpredictable Needs: Ideal for scenarios where the exact files or code snippets needed are unknown at the start.
  • Use Case: When an agent needs to refactor a complex feature, this Skill can help it discover all relevant files, understand terminology, and identify missing pieces of information step-by-step.

Quick Start

Use the iterative-retrieval skill to find relevant files for the task 'Implement user authentication flow'.

Frequently Asked Questions about iterative-retrieval

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

FAQPage Schema
How do I gather codebase context for a multi-agent workflow when the required files are unknown?

Progressive context retrieval solves this by dispatching initial queries, evaluating relevance, and refining search criteria up to three times to dynamically gather necessary codebase context for subagents.

What is the best way to refine context for complex agent tasks in a codebase?

Iterative search refines context for complex agent tasks by progressively evaluating retrieved content and adjusting criteria, looping up to three times to ensure all relevant files and snippets are discovered.

How does iterative search handle unpredictable context needs during codebase exploration?

Iterative search handles unpredictable context needs by dispatching queries, evaluating the retrieved content for relevance, and refining search criteria across multiple loops to discover missing information step-by-step.

Can I use progressive context retrieval to find all relevant files for a complex feature refactor?

Yes, progressive context retrieval is ideal for refactoring complex features because it iteratively searches and evaluates files to build relevant context, identifying missing pieces and understanding terminology dynamically.

Why does my subagent fail to start tasks without dynamic context gathering?

Subagents fail because they require codebase context that cannot be predicted upfront; dynamic context gathering solves this by progressively searching and evaluating files to build the necessary context.

What are the limitations of iterative context retrieval in agentic systems?

The iterative context retrieval pattern is limited to looping up to three times to evaluate and refine search criteria, meaning highly fragmented codebases might still lack full context coverage after the maximum iterations.