iterative-retrieval

原文涵盖的主题

• Iterative Retrieval Pattern
• When to Activate
• The Problem
• The Solution: Iterative Retrieval
• Phase 1: DISPATCH
• Phase 2: EVALUATE
• Phase 3: REFINE
• Phase 4: LOOP

Solves the "context problem" in multi-agent workflows where subagents don't know what context they need until they start working. Spawning subagents that need codebase context they cannot predict upfront Building multi-agent workflows where context is progressively refined Encountering "context too large" or "missing context" failures in agent tasks Designing RAG-like retrieval pipelines for code exploration Optimizing token usage in agent orchestration Subagents are spawned with limited context. They don't know: Which files contain relevant code What patterns exist in the codebase What terminology the project uses Standard approaches fail: Send everything: Exceeds context limits Send nothing: Agent lacks critical information Guess what's needed: Often wrong A 4-phase loop that progressively refines context: Initial broad query to gather candidate files: Assess retrieved content for relevance: Scoring criteria: High (0.8-1.0): Directly implements target functionality Medium (0.5-0.7): Contains related patterns or types Low (0.2-0.4): Tangentially related None (0-0.2): Not relevant, exclude Update search criteria based on evaluation: Repeat with refined criteria (max 3 cycles): Use

这里只展示节选，请以完整原文和当前要求为准。