codex-clilisted

OpenAI Codex CLI orchestration for AI-assisted development using gpt-5.3-codex model family. Model variants: gpt-5.3-codex (medium), gpt-5.3-codex-high, gpt-5.3-codex-xhigh. Capabilities: code generation, refactoring, automated editing, parallel task execution, session management, code review, architecture analysis, and MCP integration. Actions: analyze, implement, review, fix, refactor with Codex. Keywords: Codex CLI, gpt-5.3-codex, codex exec, code generation, refactoring, parallel execution, session resume, code review, second opinion, independent review, architecture validation, Context7 MCP. Use when: delegating complex code tasks to Codex, running multi-agent workflows, executing automated reviews, implementing features with AI assistance, resuming previous sessions, querying OpenAI documentation. Triggers: 'use codex', 'codex exec', 'run with codex', 'codex resume', 'implement with codex', 'review with codex', 'codex docs'.
alfredolopez80/multi-agent-ralph-loop · ★ 146 · AI & Automation · score 74

Install: claude install-skill alfredolopez80/multi-agent-ralph-loop

# Codex CLI Integration Skill (v2.37) ## v2.89 Key Changes (GPT-5.3-CODEX FAMILY) - **Model family**: `gpt-5.3-codex` with three reasoning tiers - **Reasoning tiers**: medium (default), high, xhigh (maximum) - **Profiles**: Pre-configured profiles for each reasoning level - **Backward compatible**: Still supports model override via `-m` flag **Reasoning Tier Selection**: | Tier | Model | Use Case | Command | |------|-------|----------|---------| | `medium` | `gpt-5.3-codex` | Default, fast code tasks | `codex exec "prompt"` | | `high` | `gpt-5.3-codex-high` | Complex analysis, security | `codex exec --profile high "prompt"` | | `xhigh` | `gpt-5.3-codex-xhigh` | Architecture, critical review | `codex exec --profile xhigh "prompt"` | **ultrathink** - Take a deep breath. We're not here to write code. We're here to make a dent in the universe. ## The Vision Codex orchestration should feel inevitable: minimal risk, maximum clarity. ## Your Work, Step by Step 1. **Select strategy**: Model, sandbox, and reasoning effort. 2. **Prepare context**: Inject the smallest, sharpest prompt. 3. **Execute**: Run Codex with clear constraints. 4. **Validate output**: Check for correctness and scope compliance. 5. **Summarize**: Report findings and next steps. ## Ultrathink Principles in Practice - **Think Different**: Choose the safest path to insight. - **Obsess Over Details**: Respect sandbox boundaries. - **Plan Like Da Vinci**: Shape the prompt before execution. - **Craft, Don't Code