القائمة
Batch-process directory-instructions.json to generate AGENTS.md files in parallel
يُحدَّد Prompt الذي يراجع المصدر أولًا بشكل افتراضي. يمكنك التبديل إلى أمر مباشر أو تنزيل نسخة محلية.
اقرأ SKILL.md وأي ملفات مرافقة يعرضها SkillsMP قبل أن تقرر التثبيت.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
يتجاوز الأمر المباشر Prompt المخصّص للمراجعة. افحص المصدر قبل تشغيله.
npx skills add https://github.com/devclarityai/environment-setup --skill process-directory-listيبقى الأمر في سطر واحد. مرّر أفقيًا لمراجعته كاملًا قبل النسخ.
تفضّل نسخة محلية؟ نزّل الملفات المتاحة حاليًا لدى SkillsMP.
عرض SKILL.md
استنادا إلى تصنيف SOC المهني
| name | process-directory-list |
| description | Batch-process directory-instructions.json to generate AGENTS.md files in parallel |
Process the directory-instructions.json file and generate AGENTS.md files
for each directory by invoking the /generate-instructions skill.
This skill orchestrates the batch generation of AGENTS.md files for multiple
directories in the monorepo. It reads the directory list JSON file and
systematically processes each directory item.
Reads from directory-instructions.json in the project root, which has this
structure:
{
"items": [
{
"id": "1",
"directory": "app/controllers",
"output_file": "AGENTS.md",
"status": "pending",
"analyzed_at": null
}
directory-instructions.json from the current working directoryitems arrayFor each item in the items array where status is "pending":
Use Codex agent delegation to process each pending directory in parallel.
Each agent receives this prompt:
Process directory "{directory_path}" from directory-instructions.json
(ID: {id}):
Steps:
1. Read directory-instructions.json
2. Update item {id} to set status="in_progress"
3. Write the updated JSON back
4. Run: /generate-instructions {directory_path}
5. Read directory-instructions.json again
6. Update item {id} to set status="completed" and
analyzed_at="{ISO timestamp}"
7. Write the final JSON back
8. If any errors occur, set status="error" instead
The /generate-instructions skill will analyze the directory and create the
AGENTS.md file.
CRITICAL: Spawn all agents in a single batch so each directory is processed concurrently.
Example:
spawn_agent(worker, "Process app/controllers and generate AGENTS.md")
spawn_agent(worker, "Process app/models and generate AGENTS.md")
spawn_agent(worker, "Process app/services and generate AGENTS.md")
... (one agent per pending directory)
This allows all directories to be processed simultaneously.
After all agents have completed, display a summary:
Completed: 6 directories
Errors: 1 directory
Skipped: 1 directory (already completed)
Launched 7 agents in parallel
pending
items# 1. First create the directory list
/create-directory-list
# 2. Then process all directories
/process-directory-list
# 3. Check results
cat directory-instructions.json
/generate-instructions directly; do not reimplement
its logic