From agent-codex
Use this skill when delegating complex coding tasks to OpenAI Codex CLI and coordinating work between Claude and Codex.
npx claudepluginhub mustafaakben/codex-agent-plugin --plugin agent-codexThis skill uses the workspace's default tool permissions.
This skill defines delegation patterns for Codex-assisted coding.
Creates isolated Git worktrees for feature branches with prioritized directory selection, gitignore safety checks, auto project setup for Node/Python/Rust/Go, and baseline verification.
Executes implementation plans in current session by dispatching fresh subagents per independent task, with two-stage reviews: spec compliance then code quality.
Dispatches parallel agents to independently tackle 2+ tasks like separate test failures or subsystems without shared state or dependencies.
This skill defines delegation patterns for Codex-assisted coding.
Before Codex operations:
.codex/config.toml.skills/codex-ecosystem/SKILL.md guidance after confirmation.model = "gpt-5.4"
model_reasoning_effort = "xhigh"
sandbox_mode = "workspace-write"
approval_policy = "on-request"
Delegate when tasks are multi-file, long-running, architecture-heavy, or require deep analysis.
Use mcp__codex-native__codex and mcp__codex-native__codex-reply for interactive multi-turn work.
Use codex exec for one-off scripted tasks.
Use background execution for long tasks and summarize results after completion.
threadId should be preserved for follow-up calls.scripts/codex-interactive.py is process-local and in-memory.gpt-5.4 + low reasoninggpt-5.4 + medium reasoninggpt-5.4 + high reasoninggpt-5.4 + xhigh reasoningWhen crafting prompts for Codex, follow the GPT-5.4 prompt patterns in references/codex-prompt-guidance.md. Key patterns:
references/codex-cli-reference.md — CLI commands, flags, configreferences/codex-prompt-guidance.md — GPT-5.4 prompt patterns and best practicesexamples/delegation-patterns.md — delegation examplesskills/codex-ecosystem/SKILL.md — bootstrap and MCP setup