القائمة
Delegate independent subtasks to `claude -p` instances. Use when you need to parallelize work across multiple files or modules that have no write conflicts.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
Use when reviewing code from a test engineer's perspective. Systematically decompose requirements into test scenarios, map them to code paths, hunt coverage gaps, and force the developer to fix or disclose every hole before delivery. If a real test engineer could find a bug you missed, you failed; if you fabricate a gap that wastes the developer's time, you failed too.
Use when reviewing embedded or firmware code changes, especially in C/C++, bare-metal, RTOS, driver, ISR, DMA, boot, NFC, or other hardware-facing paths where cross-review can catch correctness, safety, and architecture-coupling issues
Server-side automated bug fix triggered by n8n. Analyzes ZenTao bug, fixes code, commits, writes Issue/MR description files.
Use when the user asks to fix a bug, resolve an issue, or provides a bug URL/bug ID from 禅道, GitLab, GitHub, Jira, or similar systems; especially when the work needs a full workflow of reading the bug, fixing code, verifying, creating issue/MR, and writing status back to the tracker.
698/645 电表协议串口发帧与解析 Skill，支持组帧、发送、接收、解析和断言验证，用于修bug后快速回归验证
Switch SOUL.md between normal and evil personas. Triggers: /soul_good, /soul_evil, /soul_status
استنادا إلى تصنيف SOC المهني
| name | claude-code-delegation |
| description | Delegate independent subtasks to `claude -p` instances. Use when you need to parallelize work across multiple files or modules that have no write conflicts. |
--dangerously-skip-permissions only skips internal promptsclaude -p cannot reach the API, run it outside the restrictive sandbox or escalateWorking directory: <absolute-path>
Target files: <file1>, <file2>
Task: <clear description>
Input: <context or code snippets>
Requirements:
- Modify only the listed files
- Run verification: <command>
- Report: summary of changes + verification output
- End with: TASK_COMPLETE or TASK_FAILED: <reason>
| Strategy | When to use |
|---|---|
| File locks | Pre-declare locks before delegating |
| Phased execution | Stage 1: interfaces; Stage 2: parallel implementations |
| Isolation | Separate directories/workspaces for independent work |