Skip to content

borghei/Claude-Skills / engineering/codex-cli-specialist

codex-cli-specialist

Unverified

Converts Claude Code skills to OpenAI Codex format (agents/openai.yaml), checks dual-platform compatibility and builds a skills index. Use when porting or syncing skills to Codex, or setting up Codex CLI for a skill library.

assets/openai-yaml-template.yaml

View on GitHub
# =============================================================================
# agents/openai.yaml - Codex CLI Skill Configuration Template
# =============================================================================
#
# This template defines a skill for OpenAI Codex CLI.
# Copy this file to your-skill/agents/openai.yaml and customize it.
#
# Documentation: references/codex-cli-guide.md
# Converter tool: python scripts/codex_skill_converter.py
# Validator tool: python scripts/cross_platform_validator.py
#
# =============================================================================

# ---------------------------------------------------------------------------
# REQUIRED: Skill identity
# ---------------------------------------------------------------------------

# Unique identifier for the skill (kebab-case, lowercase)
name: my-skill-name

# Short description for skill discovery and matching.
# Codex uses this to decide when to activate the skill.
# Be specific about what the skill does and when to use it.
description: >
  Expert guidance for [DOMAIN]. Analyzes [X], generates [Y],
  and enforces [Z] best practices. Use when working with [CONTEXT].

# ---------------------------------------------------------------------------
# RECOMMENDED: Behavioral instructions
# ---------------------------------------------------------------------------

# Instructions define how the agent behaves when this skill is active.
# Write in second person ("You are...") with clear structure.
# Keep under 2000 words for optimal context usage.
instructions: |
  You are a senior [DOMAIN] specialist with deep expertise in [AREAS].

  ## Core Responsibilities
  - [Responsibility 1: what you help with]
  - [Responsibility 2: what you analyze]
  - [Responsibility 3: what you generate]

  ## Process
  When the user asks about [PRIMARY TASK]:
  1. Understand the context and requirements
  2. Run the appropriate analysis tool
  3. Apply best practices from references/
  4. Provide structured recommendations

  When the user asks about [SECONDARY TASK]:
  1. [Step 1]
  2. [Step 2]
  3. [Step 3]

  ## Quality Standards
  - Always explain your reasoning
  - Provide concrete, actionable recommendations
  - Reference specific files and line numbers when relevant
  - Use the available tools for data-driven analysis
  - Follow the patterns in references/ for domain best practices

  ## Output Format
  Structure your responses with:
  - A brief summary of findings
  - Detailed analysis with evidence
  - Prioritized recommendations
  - Next steps for the user

# ---------------------------------------------------------------------------
# RECOMMENDED: Tool definitions
# ---------------------------------------------------------------------------

# Tools expose Python scripts to the Codex agent.
# Each tool maps to a command that Codex can execute.
tools:
  # Tool 1: Primary analysis tool
  - name: primary_analyzer
    description: >
      Analyzes [X] and produces a structured assessment report.
      Use this when the user asks to review, audit, or evaluate [X].
    command: python scripts/primary_analyzer.py
    # Optional: define accepted arguments
    # args:
    #   - name: input_path
    #     description: Path to the file or directory to analyze
    #     required: true
    #   - name: format
    #     description: Output format (json or text)
    #     required: false
    #     default: text

  # Tool 2: Generator tool
  - name: artifact_generator
    description: >
      Generates [Y] artifacts from templates and configuration.
      Use this when the user asks to create, scaffold, or produce [Y].
    command: python scripts/artifact_generator.py

  # Tool 3: Validator tool (optional)
  # - name: validator
  #   description: >
  #     Validates [Z] against best practices and standards.
  #     Use this when the user asks to check, verify, or validate [Z].
  #   command: python scripts/validator.py

# ---------------------------------------------------------------------------
# OPTIONAL: Model and version
# ---------------------------------------------------------------------------

# Preferred model for this skill (optional). Omit it so the skill inherits the
# user's CLI default; pin an id from OpenAI's current model list only if needed.
# model: <model-id>

# Skill version (follows semver)
version: 1.0.0
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 322kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 322kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 322kUpdated Claude CodeCodexVerifiedInstall