Batch Worker — 100 AI Agents in Parallel, Zero-Token Cleanup

Batch Worker — 百个 AI 智能体并行处理，零 Token 成本清理

Batch Worker: 100 AI Agents Running in Parallel Batch Worker：百个 AI 智能体并行运行

The Problem: Auditing a codebase takes hours when you go file by file. Content creation, search, fixes — every task is bottlenecked by sequential execution. 问题所在：逐个文件审计代码库需要耗费数小时。无论是内容创作、搜索还是修复，每一项任务都受限于顺序执行的瓶颈。

The Solution: Batch Worker is an OpenClaw skill that dispatches up to 100 AI agents in parallel with staggered launch to avoid rate limits. 解决方案：Batch Worker 是一项 OpenClaw 技能，它能够并行调度多达 100 个 AI 智能体，并通过错峰启动来规避速率限制。

Three-Step Pipeline 三步流水线

1. ai_planner -> Analyze project, generate audit plan with 100 soldier prompts

2. ai_planner -> 分析项目，生成包含 100 个“士兵”提示词的审计计划

3. core_taskPipeline -> Dispatch 100 agents in staggered batches (10/batch x 20ms)

4. core_taskPipeline -> 以错峰批次调度 100 个智能体（每批 10 个，间隔 20 毫秒）

5. ai_collector -> Collect reports, deduplicate, rank by severity — zero LLM tokens

6. ai_collector -> 收集报告、去重、按严重程度排序——全程零 LLM Token 消耗

104 Audit Dimensions 104 项审计维度

• Domain Dimensions: Security 42 (injection, XSS, CSRF, auth, encryption…), Architecture 12 (boundaries, cycles, idempotency, resilience), Performance 12 (N+1, caching, memory leaks, lock contention), Code Quality 10 (complexity, error handling, dead code), Language-specific 10 (Promise, async, EventEmitter, Stream), DevOps 8 (observability, CI/CD, config, containers), Compliance 4 (data privacy, audit trail, accessibility)
• 领域维度：安全 42 项（注入、XSS、CSRF、认证、加密等）、架构 12 项（边界、循环、幂等性、弹性）、性能 12 项（N+1 问题、缓存、内存泄漏、锁竞争）、代码质量 10 项（复杂度、错误处理、死代码）、语言特性 10 项（Promise、异步、EventEmitter、流）、DevOps 8 项（可观测性、CI/CD、配置、容器）、合规性 4 项（数据隐私、审计追踪、可访问性）

83 Task Types 83 种任务类型

From code audit to content creation, search to fix, translation to analysis — batch-worker handles them all. 从代码审计到内容创作，从搜索到修复，从翻译到分析——Batch-worker 都能胜任。

Zero-Token Cleanup 零 Token 清理

After 100 agents finish, ai_collector automatically extracts JSON findings, deduplicates, and merges — using zero LLM tokens. Pure script, no hallucination. 在 100 个智能体完成工作后，ai_collector 会自动提取 JSON 格式的发现结果，进行去重和合并——全程不消耗任何 LLM Token。纯脚本处理，无幻觉。

GitHub: https://github.com/haoyun18881-beep/batch-worker Docs: https://haoyun18881-beep.github.io/batch-worker/