Back to Details

iterative-retrieval

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

Iterative Retrieval Pattern

迭代式检索模式

Solves the "context problem" in multi-agent workflows where subagents don't know what context they need until they start working.
解决多Agent工作流中的“上下文问题”——子Agent在开始工作前不知道自己需要什么上下文。

The Problem

问题

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
子Agent启动时只有有限的上下文，它们不知道：
  • 哪些文件包含相关代码
  • 代码库中存在哪些模式
  • 项目使用什么术语
标准方法存在缺陷：
  • 发送全部内容：超出上下文限制
  • 不发送任何内容：Agent缺乏关键信息
  • 猜测所需内容：通常不准确

The Solution: Iterative Retrieval

解决方案：迭代式检索

A 4-phase loop that progressively refines context:
┌─────────────────────────────────────────────┐
│                                             │
│   ┌──────────┐      ┌──────────┐            │
│   │ DISPATCH │─────▶│ EVALUATE │            │
│   └──────────┘      └──────────┘            │
│        ▲                  │                 │
│        │                  ▼                 │
│   ┌──────────┐      ┌──────────┐            │
│   │   LOOP   │◀─────│  REFINE  │            │
│   └──────────┘      └──────────┘            │
│                                             │
│        Max 3 cycles, then proceed           │
└─────────────────────────────────────────────┘
一个分4阶段的循环，逐步优化上下文：
┌─────────────────────────────────────────────┐
│                                             │
│   ┌──────────┐      ┌──────────┐            │
│   │ DISPATCH │─────▶│ EVALUATE │            │
│   └──────────┘      └──────────┘            │
│        ▲                  │                 │
│        │                  ▼                 │
│   ┌──────────┐      ┌──────────┐            │
│   │   LOOP   │◀─────│  REFINE  │            │
│   └──────────┘      └──────────┘            │
│                                             │
│        Max 3 cycles, then proceed           │
└─────────────────────────────────────────────┘

Phase 1: DISPATCH

阶段1：DISPATCH（分发）

Initial broad query to gather candidate files:
javascript
// Start with high-level intent
const initialQuery = {
  patterns: ['src/**/*.ts', 'lib/**/*.ts'],
  keywords: ['authentication', 'user', 'session'],
  excludes: ['*.test.ts', '*.spec.ts']
};

// Dispatch to retrieval agent
const candidates = await retrieveFiles(initialQuery);
初始宽泛查询以收集候选文件：
javascript
// Start with high-level intent
const initialQuery = {
  patterns: ['src/**/*.ts', 'lib/**/*.ts'],
  keywords: ['authentication', 'user', 'session'],
  excludes: ['*.test.ts', '*.spec.ts']
};

// Dispatch to retrieval agent
const candidates = await retrieveFiles(initialQuery);

Phase 2: EVALUATE

阶段2：EVALUATE（评估）

Assess retrieved content for relevance:
javascript
function evaluateRelevance(files, task) {
  return files.map(file => ({
    path: file.path,
    relevance: scoreRelevance(file.content, task),
    reason: explainRelevance(file.content, task),
    missingContext: identifyGaps(file.content, task)
  }));
}
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
评估检索到的内容的相关性：
javascript
function evaluateRelevance(files, task) {
  return files.map(file => ({
    path: file.path,
    relevance: scoreRelevance(file.content, task),
    reason: explainRelevance(file.content, task),
    missingContext: identifyGaps(file.content, task)
  }));
}
评分标准：
  • 高（0.8-1.0）：直接实现目标功能
  • 中（0.5-0.7）：包含相关模式或类型
  • 低（0.2-0.4）：间接相关
  • 无（0-0.2）：不相关，排除

Phase 3: REFINE

阶段3：REFINE（优化）

Update search criteria based on evaluation:
javascript
function refineQuery(evaluation, previousQuery) {
  return {
    // Add new patterns discovered in high-relevance files
    patterns: [...previousQuery.patterns, ...extractPatterns(evaluation)],

    // Add terminology found in codebase
    keywords: [...previousQuery.keywords, ...extractKeywords(evaluation)],

    // Exclude confirmed irrelevant paths
    excludes: [...previousQuery.excludes, ...evaluation
      .filter(e => e.relevance < 0.2)
      .map(e => e.path)
    ],

    // Target specific gaps
    focusAreas: evaluation
      .flatMap(e => e.missingContext)
      .filter(unique)
  };
}
基于评估结果更新搜索条件：
javascript
function refineQuery(evaluation, previousQuery) {
  return {
    // Add new patterns discovered in high-relevance files
    patterns: [...previousQuery.patterns, ...extractPatterns(evaluation)],

    // Add terminology found in codebase
    keywords: [...previousQuery.keywords, ...extractKeywords(evaluation)],

    // Exclude confirmed irrelevant paths
    excludes: [...previousQuery.excludes, ...evaluation
      .filter(e => e.relevance < 0.2)
      .map(e => e.path)
    ],

    // Target specific gaps
    focusAreas: evaluation
      .flatMap(e => e.missingContext)
      .filter(unique)
  };
}

Phase 4: LOOP

阶段4：LOOP（循环）

Repeat with refined criteria (max 3 cycles):
javascript
async function iterativeRetrieve(task, maxCycles = 3) {
  let query = createInitialQuery(task);
  let bestContext = [];

  for (let cycle = 0; cycle < maxCycles; cycle++) {
    const candidates = await retrieveFiles(query);
    const evaluation = evaluateRelevance(candidates, task);

    // Check if we have sufficient context
    const highRelevance = evaluation.filter(e => e.relevance >= 0.7);
    if (highRelevance.length >= 3 && !hasCriticalGaps(evaluation)) {
      return highRelevance;
    }

    // Refine and continue
    query = refineQuery(evaluation, query);
    bestContext = mergeContext(bestContext, highRelevance);
  }

  return bestContext;
}
使用优化后的条件重复执行（最多3轮）：
javascript
async function iterativeRetrieve(task, maxCycles = 3) {
  let query = createInitialQuery(task);
  let bestContext = [];

  for (let cycle = 0; cycle < maxCycles; cycle++) {
    const candidates = await retrieveFiles(query);
    const evaluation = evaluateRelevance(candidates, task);

    // Check if we have sufficient context
    const highRelevance = evaluation.filter(e => e.relevance >= 0.7);
    if (highRelevance.length >= 3 && !hasCriticalGaps(evaluation)) {
      return highRelevance;
    }

    // Refine and continue
    query = refineQuery(evaluation, query);
    bestContext = mergeContext(bestContext, highRelevance);
  }

  return bestContext;
}

Practical Examples

实际示例

Example 1: Bug Fix Context

示例1：Bug修复上下文

Task: "Fix the authentication token expiry bug"

Cycle 1:
  DISPATCH: Search for "token", "auth", "expiry" in src/**
  EVALUATE: Found auth.ts (0.9), tokens.ts (0.8), user.ts (0.3)
  REFINE: Add "refresh", "jwt" keywords; exclude user.ts

Cycle 2:
  DISPATCH: Search refined terms
  EVALUATE: Found session-manager.ts (0.95), jwt-utils.ts (0.85)
  REFINE: Sufficient context (2 high-relevance files)

Result: auth.ts, tokens.ts, session-manager.ts, jwt-utils.ts
Task: "Fix the authentication token expiry bug"

Cycle 1:
  DISPATCH: Search for "token", "auth", "expiry" in src/**
  EVALUATE: Found auth.ts (0.9), tokens.ts (0.8), user.ts (0.3)
  REFINE: Add "refresh", "jwt" keywords; exclude user.ts

Cycle 2:
  DISPATCH: Search refined terms
  EVALUATE: Found session-manager.ts (0.95), jwt-utils.ts (0.85)
  REFINE: Sufficient context (2 high-relevance files)

Result: auth.ts, tokens.ts, session-manager.ts, jwt-utils.ts

Example 2: Feature Implementation

示例2：功能实现

Task: "Add rate limiting to API endpoints"

Cycle 1:
  DISPATCH: Search "rate", "limit", "api" in routes/**
  EVALUATE: No matches - codebase uses "throttle" terminology
  REFINE: Add "throttle", "middleware" keywords

Cycle 2:
  DISPATCH: Search refined terms
  EVALUATE: Found throttle.ts (0.9), middleware/index.ts (0.7)
  REFINE: Need router patterns

Cycle 3:
  DISPATCH: Search "router", "express" patterns
  EVALUATE: Found router-setup.ts (0.8)
  REFINE: Sufficient context

Result: throttle.ts, middleware/index.ts, router-setup.ts
Task: "Add rate limiting to API endpoints"

Cycle 1:
  DISPATCH: Search "rate", "limit", "api" in routes/**
  EVALUATE: No matches - codebase uses "throttle" terminology
  REFINE: Add "throttle", "middleware" keywords

Cycle 2:
  DISPATCH: Search refined terms
  EVALUATE: Found throttle.ts (0.9), middleware/index.ts (0.7)
  REFINE: Need router patterns

Cycle 3:
  DISPATCH: Search "router", "express" patterns
  EVALUATE: Found router-setup.ts (0.8)
  REFINE: Sufficient context

Result: throttle.ts, middleware/index.ts, router-setup.ts

Integration with Agents

与Agent集成

Use in agent prompts:
markdown
When retrieving context for this task:
1. Start with broad keyword search
2. Evaluate each file's relevance (0-1 scale)
3. Identify what context is still missing
4. Refine search criteria and repeat (max 3 cycles)
5. Return files with relevance >= 0.7
在Agent提示词中使用：
markdown
When retrieving context for this task:
1. Start with broad keyword search
2. Evaluate each file's relevance (0-1 scale)
3. Identify what context is still missing
4. Refine search criteria and repeat (max 3 cycles)
5. Return files with relevance >= 0.7

Best Practices

最佳实践

  1. Start broad, narrow progressively - Don't over-specify initial queries
  2. Learn codebase terminology - First cycle often reveals naming conventions
  3. Track what's missing - Explicit gap identification drives refinement
  4. Stop at "good enough" - 3 high-relevance files beats 10 mediocre ones
  5. Exclude confidently - Low-relevance files won't become relevant
  1. 从宽泛开始，逐步缩小范围——不要过度指定初始查询
  2. 学习代码库术语——第一轮通常会揭示命名约定
  3. 跟踪缺失内容——明确的差距识别推动优化
  4. 适可而止——3个高相关性文件优于10个中等相关性文件
  5. 果断排除无关内容——低相关性文件不会变得相关

Related

相关内容

  • The Longform Guide - Subagent orchestration section
  • continuous-learning
    skill - For patterns that improve over time
  • Agent definitions in 
    ~/.claude/agents/
  • The Longform Guide——子Agent编排部分
  • continuous-learning
    skill——适用于随时间改进的模式
  • Agent定义位于
    ~/.claude/agents/