codex-implement
Delegate complex implementation tasks to Codex. Use when user says codex implement, delegate implementation, or has a complex coding task.
Best use case
codex-implement is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
Delegate complex implementation tasks to Codex. Use when user says codex implement, delegate implementation, or has a complex coding task.
Teams using codex-implement should expect a more consistent output, faster repeated execution, less prompt rewriting.
When to use this skill
- You want a reusable workflow that can be run more than once with consistent structure.
When not to use this skill
- You only need a quick one-off answer and do not need a reusable workflow.
- You cannot install or maintain the underlying files, dependencies, or repository context.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/codex-implement/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How codex-implement Compares
| Feature / Agent | codex-implement | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
Delegate complex implementation tasks to Codex. Use when user says codex implement, delegate implementation, or has a complex coding task.
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
Related Guides
SKILL.md Source
# Codex Implement Skill
Delegate complex implementation tasks to the Codex agent for high-quality code.
## When to Use
- Complex multi-file implementations
- Algorithms requiring careful design
- User explicitly wants Codex to implement
## Steps
Before any file operations, resolve the `.agent-collab` directory so commands work outside the project root:
```bash
AGENT_COLLAB_DIR="${AGENT_COLLAB_DIR:-}"
if [ -n "$AGENT_COLLAB_DIR" ]; then
if [ -d "$AGENT_COLLAB_DIR/.agent-collab" ]; then
AGENT_COLLAB_DIR="$AGENT_COLLAB_DIR/.agent-collab"
elif [ ! -d "$AGENT_COLLAB_DIR" ]; then
AGENT_COLLAB_DIR=""
fi
fi
if [ -z "$AGENT_COLLAB_DIR" ]; then
AGENT_COLLAB_DIR="$(pwd)"
while [ "$AGENT_COLLAB_DIR" != "/" ] && [ ! -d "$AGENT_COLLAB_DIR/.agent-collab" ]; do
AGENT_COLLAB_DIR="$(dirname "$AGENT_COLLAB_DIR")"
done
AGENT_COLLAB_DIR="$AGENT_COLLAB_DIR/.agent-collab"
fi
```
If `$AGENT_COLLAB_DIR` does not exist, stop and ask for the project root.
### 1. Gather Requirements
From user, collect:
- What needs implementing
- Target file paths
- Constraints and patterns to follow
- Dependencies and interfaces
### 2. Create Implementation Spec
Write to `$AGENT_COLLAB_DIR/requests/task.md`:
```markdown
# Task Request for Codex
## Task Type: IMPLEMENT
## Timestamp
[Current timestamp]
## Implementation Request
[Detailed description]
## Target Files
- Primary: [main file path]
- Secondary: [supporting files]
## Requirements
1. [Requirement 1]
2. [Requirement 2]
...
## Interfaces & Contracts
[Interfaces the code must satisfy]
## Existing Code Context
[Relevant existing code to integrate with]
## Patterns to Follow
[Reference existing patterns]
## Constraints
- [List constraints]
```
### 3. Update Status
Write `pending` to `$AGENT_COLLAB_DIR/status`
### 4. Trigger Codex
```bash
tmux send-keys -t 1 '$read-task' && sleep 0.5 && tmux send-keys -t 1 Enter Enter
```
### 5. Notify User
Tell user briefly that implementation was delegated to Codex.
### 6. Wait for Codex (Background Polling)
Start a background polling loop to wait for Codex to complete. Run this EXACT bash command (with `$AGENT_COLLAB_DIR/status`) using the Bash tool with `run_in_background: true`:
```bash
while [ "$(cat "$AGENT_COLLAB_DIR/status")" != "done" ]; do sleep 5; done; echo "CODEX_COMPLETE"
```
Note: Use 5 second intervals since implementations take longer.
CRITICAL: Use the resolved `$AGENT_COLLAB_DIR/status` path so polling works outside the project root. Use background execution so you can continue helping the user while waiting.
### 7. Auto-Read Response
When poll completes, automatically:
1. Read `$AGENT_COLLAB_DIR/responses/response.md`
2. Present the implementation to user
3. Ask if user wants to integrate the code
4. Reset `$AGENT_COLLAB_DIR/status` to `idle`Related Skills
implement_plan
Implement technical plans from thoughts/shared/plans with verification
task-implementation
Instructions for implementing task plans with progressive tracking and change record - Brought to you by microsoft/edge-ai Triggers on: **/.copilot-tracking/changes/*.md
openai-codex
OpenAI Codex CLI usage patterns, configuration, sandboxing, and best practices for AI-assisted development.
julien-workflow-advice-codex
Get OpenAI Codex CLI's opinion on code, bugs, or implementation. Use when you want a second AI perspective during coding sessions.
implementing-android-code
This skill should be used when implementing Android code in Bitwarden. Covers critical patterns, gotchas, and anti-patterns unique to this codebase. Triggered by "How do I implement a ViewModel?", "Create a new screen", "Add navigation", "Write a repository", "BaseViewModel pattern", "State-Action-Event", "type-safe navigation", "@Serializable route", "SavedStateHandle persistence", "process death recovery", "handleAction", "sendAction", "Hilt module", "Repository pattern", "implementing a screen", "adding a data source", "handling navigation", "encrypted storage", "security patterns", "Clock injection", "DataState", or any questions about implementing features, screens, ViewModels, data sources, or navigation in the Bitwarden Android app.
cqrs-implementation
Implement Command Query Responsibility Segregation for scalable architectures. Use when separating read and write models, optimizing query performance, or building event-sourced systems.
consult-codex
Compare OpenAI Codex GPT-5.2 and code-searcher responses for comprehensive dual-AI code analysis. Use when you need multiple AI perspectives on code questions.
codex-team
Use when you have 2+ tasks that Codex agents should execute. Runtime-native: Codex sub-agents when available, Codex CLI fallback otherwise. Handles file conflicts via merge/wave strategies. Triggers: "codex team", "spawn codex", "codex agents", "use codex for", "codex fix".
codex
Run OpenAI's Codex CLI agent in non-interactive mode using `codex exec`. Use when delegating coding tasks to Codex, running Codex in scripts/automation, or when needing a second agent to work on a task in parallel.
codex-sessions-skill-scan
Daily skill health scan: analyze ~/.codex/sessions plus per-repo session logs under ~/dev (default last 1 day) and summarize skill invocations + likely failures for personal skills in ~/dev/agent-skills (missing paths, tool failures, complex-task word triggers). Optional: include best-effort local OTel signals.
codex-reviewer
Use OpenAI's Codex CLI as an independent code reviewer to provide second opinions on code implementations, architectural decisions, code specifications, and pull requests. Trigger when users request code review, second opinion, independent review, architecture validation, or mention Codex review. Provides unbiased analysis using GPT-5-Codex model through the codex exec command for non-interactive reviews.
codex-review
Two-pass adversarial review of design documents and implementation plans using OpenAI Codex CLI. Invokes Codex to review plans section-by-section (pass 1), then holistically (pass 2), feeding critique back for revision. Use when you have a design doc, architecture plan, or implementation plan that should be stress-tested before execution.