القائمة
Use when creating, updating, auditing, or validating Agent Skills. Provides validation scripts and guidance for keeping skills accurate and compliant with the Agent Skills specification.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
Create a subsystem orientation skill after deep research on a stable, non-churning part of the codebase when no skill exists for it. Use when you have just read many files, traced call chains, or explored architecture to understand how a subsystem works — and that knowledge should be captured to prevent the same context burn on every future session touching that area. Triggers include: had to explore 5+ files to understand one concept, traced IPC/process/worker/pipeline topology manually, or answered "how does X actually work" from scratch.
Use when executing implementation plans produced by the feature-planning skill. Orchestrates execution subagents (one plan phase at a time), dispatches review subagents for thorough quality enforcement after each plan, and manages fix cycles when review finds issues. Trigger when: (1) the user explicitly says "execute the plans", "implement the feature", or "work through the plans"; OR (2) the user asks to start implementation and every plan for the single feature being executed (all lettered plans A–Z for that specific feature slug, covering its full scope) exists and is schema-valid in artifacts/plans/pending/TASK-*-{A..Z}-*.md. Not for single-plan execution — use plan_complete_step directly for those.
Use when decomposing a major feature design into dependency-ordered implementation plans. Handles the full pipeline from design document to validated, cross-referenced plan files with minimal drift. Trigger when **any** of the following are true, with user requests (4) taking priority over the other conditions: (1) the feature involves 3+ implementation parts, (2) it spans multiple architectural layers, (3) it requires coordination across multiple sessions, or (4) the user explicitly asks to break down a design doc into implementation plans. Not for single plans or simple tasks — use the Plan subagent directly for those.
Use when creating or updating VS Code Copilot agent hook files in .github/hooks. Covers all 8 hook event types (PreToolUse, PostToolUse, SessionStart, SubagentStart, SubagentStop, Stop, UserPromptSubmit, PreCompact), the stdin/stdout JSON contract, permissionDecision allow/deny/ask, exit code semantics, hook command properties, and repository conventions for Python validator scripts.
Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Claude's capabilities with specialized knowledge, workflows, or tool integrations.
Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK).
| name | skill-maintenance |
| description | Use when creating, updating, auditing, or validating Agent Skills. Provides validation scripts and guidance for keeping skills accurate and compliant with the Agent Skills specification. |
When to use: When creating new skills, updating outdated skills, or auditing skill quality.
Purpose: Check skills for format compliance and invalid references.
Usage:
# Validate all skills
python scripts/validate_skills.py
# Validate one skill
python scripts/validate_skills.py layer-helpers
# JSON output
python scripts/validate_skills.py --format=json
# Also check if code references exist
python scripts/validate_skills.py --check-refs
Checks performed:
---name, descriptionname matches directory name, lowercase with hyphensname ≤ 64 chars, no consecutive hyphensdescription ≤ 1024 chars, non-empty--check-refs).github/skills/<skill-name>/SKILL.md with frontmatter:---
name: skill-name
description: Use when [trigger conditions]. Provides [capabilities].
---
# Skill Title
**When to use:** [Clear trigger description]
---
## Section 1
Instructions...
---
## Validation Checklist
- [ ] Check 1 **→ Consequence**
- [ ] Check 2 **→ Consequence**
python scripts/validate_skills.py skill-nameWhen: Architecture changed, code moved, or skill guidance is stale.
Process:
python scripts/validate_skills.py <skill-name> --check-refsRules:
| Field | Required | Constraints |
|---|---|---|
name | Yes | 1-64 chars, lowercase alphanumeric + hyphens, must match directory |
description | Yes | 1-1024 chars, describe WHAT and WHEN |
--)Good:
description: Use when creating or modifying code in nomarr/helpers/. Helpers are pure utilities with NO nomarr.* imports.
Poor:
description: Helper layer rules.
references/)--- separatorsCode references in skills can become stale when code moves.
# Find invalid references
python scripts/validate_skills.py --check-refs
# Check specific skill
python scripts/validate_skills.py layer-components --check-refs
Fix invalid references:
Run monthly or after major refactors:
# 1. Validate all skills
python scripts/validate_skills.py --check-refs
# 2. Review any invalid references
# 3. Update stale skills
# 4. Re-validate
python scripts/validate_skills.py
Nomarr skills are organized by purpose:
| Skill | Purpose |
|---|---|
layer-* | Architecture layer guidance |
code-discovery | Codebase exploration tools |
code-generation | Boilerplate generation |
quality-analysis | QC and linting tools |
skill-maintenance | This skill (meta) |
Before committing skill changes, verify:
name and description? → Requiredname match directory name? → Requiredname lowercase with hyphens only? → Requireddescription explain WHAT and WHEN? → Required--check-refsreferences/ if not