codexforclaude
OpenAI Codex CLI configuration guide for Claude. Use when users ask about: codex config, ~/.codex/config.toml, codex sandbox policies, codex approval modes, creating codex skills, codex mcp servers, codex model settings, local LLM providers, or integrating tools with Codex.
What this skill does
# Codex CLI Configuration Guide Configure and extend OpenAI Codex CLI for optimal development workflows. ## Quick Reference | Task | Command/Location | |------|------------------| | Config file | `~/.codex/config.toml` | | Add MCP server | `codex mcp add <name> -- <command>` | | View MCP servers | `/mcp` in TUI | | Skills location | `~/.codex/skills/` or `.codex/skills/` | | Create skill | Use `$skill-creator` in Codex | ## Core Configuration Edit `~/.codex/config.toml` for all settings: ```toml # Model settings model = "gpt-5-codex" model_reasoning_effort = "medium" # minimal|low|medium|high|xhigh # Security sandbox_mode = "workspace-write" # read-only|workspace-write|danger-full-access approval_policy = "on-failure" # untrusted|on-failure|on-request|never # Trust specific projects [projects."/path/to/project"] trust_level = "trusted" ``` **For detailed config options:** See `references/config-toml.md` ## Sandbox and Approval Policies **Sandbox modes** control file system access: - `read-only` - Can only read files, no writes - `workspace-write` - Write within project directory only - `danger-full-access` - Full system access (use carefully) **Approval policies** control command execution: - `untrusted` - Only trusted commands run without approval - `on-failure` - Auto-run, ask approval only on failures - `on-request` - Model decides when to ask - `never` - Never ask (dangerous) **For security deep dive:** See `references/sandbox-approval.md` ## Adding MCP Servers ### Via CLI (Recommended) ```bash codex mcp add context7 -- npx -y @upstash/context7-mcp codex mcp add myserver --env API_KEY=xxx -- node server.js ``` ### Via config.toml ```toml [mcp_servers.context7] command = "npx" args = ["-y", "@upstash/context7-mcp"] [mcp_servers.figma] url = "https://mcp.figma.com/mcp" bearer_token_env_var = "FIGMA_OAUTH_TOKEN" ``` **For full MCP options:** See `references/mcp-servers.md` ## Creating Codex Skills Skills extend Codex with specialized knowledge. Structure: ``` ~/.codex/skills/my-skill/ ├── SKILL.md # Required: instructions + metadata ├── scripts/ # Optional: executable code ├── references/ # Optional: detailed docs └── assets/ # Optional: templates ``` **Minimal SKILL.md:** ```yaml --- name: my-skill description: When to trigger this skill --- Instructions for Codex when using this skill. ``` **Quick start:** Ask Codex to use `$skill-creator` to bootstrap a new skill. **For skill development guide:** See `references/creating-skills.md` ## Model and Provider Settings ```toml # Use specific model model = "gpt-5-codex" # Reasoning effort (Responses API) model_reasoning_effort = "high" # minimal|low|medium|high|xhigh # Verbosity (GPT-5) model_verbosity = "medium" # low|medium|high # Use local provider (LM Studio/Ollama) model_provider = "oss" ``` **For local LLM setup:** See `references/model-providers.md` ## Feature Flags Enable experimental features in `[features]`: ```toml [features] # Stable shell_tool = true parallel = true view_image_tool = true # Beta unified_exec = true shell_snapshot = true # Experimental skills = true tui2 = true ``` ## Common Workflows ### Full Autonomous Mode ```bash codex --full-auto "implement feature X" # Equivalent to: -a on-request --sandbox workspace-write ``` ### Dangerous Bypass (CI/sandboxed environments only) ```bash codex --dangerously-bypass-approvals-and-sandbox "run tests" ``` ### Project-Specific Trust ```toml [projects."/home/user/trusted-project"] trust_level = "trusted" ``` ## Codex vs Claude Code | Aspect | Codex | Claude Code | |--------|-------|-------------| | Provider | OpenAI | Anthropic | | Config | `~/.codex/config.toml` | `~/.claude/settings.json` | | Skills | `~/.codex/skills/` | `~/.claude/skills/` + plugins | | MCP | `config.toml` sections | `.mcp.json` or plugin | | Agents | N/A | Plugin agents | | Hooks | N/A | Plugin hooks | ## Reference Files - `references/config-toml.md` - Complete config.toml reference - `references/sandbox-approval.md` - Security policies deep dive - `references/mcp-servers.md` - MCP server configuration - `references/creating-skills.md` - Skill development guide - `references/model-providers.md` - Model and provider settings
Related in AI Agents
skill-development
IncludedComprehensive meta-skill for creating, managing, validating, auditing, and distributing Claude Code skills and slash commands (unified in v2.1.3+). Provides skill templates, creation workflows, validation patterns, audit checklists, naming conventions, YAML frontmatter guidance, progressive disclosure examples, and best practices lookup. Use when creating new skills, validating existing skills, auditing skill quality, understanding skill architecture, needing skill templates, learning about YAML frontmatter requirements, progressive disclosure patterns, tool restrictions (allowed-tools), skill composition, skill naming conventions, troubleshooting skill activation issues, creating custom slash commands, configuring command frontmatter, using command arguments ($ARGUMENTS, $1, $2), bash execution in commands, file references in commands, command namespacing, plugin commands, MCP slash commands, Skill tool configuration, or deciding between skills vs slash commands. Delegates to docs-management skill for official documentation.
reprompter
IncludedTransform messy prompts into well-structured, effective prompts — single or multi-agent. Use when: "reprompt", "reprompt this", "clean up this prompt", "structure my prompt", rough text needing XML tags and best practices, "reprompter teams", "repromptception", "run with quality", "smart run", "smart agents", multi-agent tasks, audits, parallel work, anything going to agent teams. Don't use when: simple Q&A, pure chat, immediate execution-only tasks. See "Don't Use When" section for details. Outputs: Structured XML/Markdown prompt, quality score (before/after), optional team brief + per-agent sub-prompts, agent team output files. Success criteria: Single mode quality score ≥ 7/10; Repromptception per-agent prompt quality score 8+/10; all required sections present, actionable and specific.
adaptive-compaction
IncludedAdaptive add-on policy and recovery layer that decides WHEN to compact, prune, snapshot, or fork -- replacing fixed-percent auto-compaction across Claude Code, Codex, and MCP-capable hosts. Trigger on auto-compact timing or damage: "when should I compact", "is it safe to compact now or start a fresh session", "auto-compact fires too early/mid-task", "switching to an unrelated task but the window still has space", "context rot", "answers get worse the longer the session runs", "the agent forgot the plan or my decisions after it summarized", "add a layer on top that manages context without changing the agent", raising autoCompactWindow to give the policy room, or installing/tuning a cross-tool compaction policy or PreCompact hook -- even when "compaction" is never said but the problem is context-window pressure or post-summarization memory loss. Do NOT use to summarize a conversation, build RAG, write a summarization prompt (decides WHEN not HOW), or answer max-context-length trivia.
agent-skill-creator
IncludedCreate cross-platform agent skills from workflow descriptions. Activates when users ask to create an agent, automate a repetitive workflow, create a custom skill, or need advanced agent creation. Triggers on phrases like create agent for, automate workflow, create skill for, every day I have to, daily I need to, turn process into agent, need to automate, create a cross-platform skill, validate this skill, export this skill, migrate this skill. Supports single skills, multi-agent suites, transcript processing, template-based creation, interactive configuration, cross-platform export, and spec validation.
llm-wiki
IncludedUse when building or maintaining a persistent personal knowledge base (second brain) in Obsidian where an LLM incrementally ingests sources, updates entity/concept pages, maintains cross-references, and keeps a synthesis current. Triggers include "second brain", "Obsidian wiki", "personal knowledge management", "ingest this paper/article/book", "build a research wiki", "compound knowledge", "Memex", or whenever the user wants knowledge to accumulate across sessions instead of being re-derived by RAG on every query.
skill-master
IncludedAgent Skills authoring, evaluation, and optimization. Create, edit, validate, benchmark, and improve skills following the agentskills.io specification. Use when designing SKILL.md files, structuring skill folders (references, scripts, assets), ingesting external documentation into skills, running trigger evals, benchmarking skill quality, optimizing descriptions, or performing blind A/B comparisons. Keywords: agentskills.io, SKILL.md, skill authoring, eval, benchmark, trigger optimization.