A senior-level engineering mentor package for Openclaw Skills that automates full-stack development, CI/CD pipelines, and GitHub workflows.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install codesmith
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install codesmith using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is CodeSmith?
CodeSmith is a production-hardened mentor package designed to elevate Openclaw Skills from basic assistants to senior-level engineering partners. It provides a comprehensive configuration for agents that need to perform real coding work, including implementing features, managing repositories, and handling deployment pipelines. By synthesizing real-world operational wisdom, CodeSmith allows agents to operate with high levels of autonomy and technical judgment.
This package is specifically built for developers who utilize OpenClaw as a primary coding partner. It focuses on sophisticated workflows such as ACP dispatch patterns for sub-agent management and production-ready CI/CD integration. Instead of just executing commands, an agent equipped with CodeSmith understands the 'why' behind its actions, utilizing annotated configurations and trust-based progression models to ensure reliable software delivery.
CodeSmith Use Cases
- Automating complex CI/CD pipelines and hosting service deployments.
- Managing GitHub repositories using the gh CLI for autonomous pull request handling.
- Coordinating multi-agent workflows through ACP dispatch patterns for implementation tasks.
- Implementing background repository maintenance via scheduled cron jobs.
- Transitioning manual coding tasks to autonomous agent-driven development cycles.
How CodeSmith Works
- The CodeSmith package is applied via the claw-mentor-mentee skill during a scheduled ingestion cycle.
- The agent adopts the senior engineering configuration defined in AGENTS.md, which includes 17 specific annotation blocks for decision-making.
- The skill stack is initialized across three tiers, ensuring the agent has the necessary tools while excluding redundant or high-risk skills.
- Working patterns and trust progression models are established, allowing the agent to move from human-reviewed changes to autonomous operations.
- Scheduled automation is triggered through cron-patterns.json, enabling the agent to perform recurring maintenance and reporting tasks independently.
CodeSmith Setup
To deploy CodeSmith, you must have the claw-mentor-mentee skill (v2.1.0+) installed and active. Ensure your environment has the GitHub CLI configured and ACP enabled for sub-agent dispatch.
# Verify GitHub CLI configuration
gh auth status
# The skill is applied automatically during ingestion.
# To review the setup guide and adoption timeline:
cat setup-guide.md
Follow the one-at-a-time adoption timeline specified in the setup guide to safely enable cron jobs and autonomous dispatch features.
CodeSmith Data Schema & Taxonomy
CodeSmith utilizes a structured metadata taxonomy to organize its engineering intelligence and operational constraints:
| File | Data Type | Description |
|---|---|---|
AGENTS.md |
Markdown/Annotations | Explains the logic behind agent configuration decisions. |
working-patterns.md |
Documentation | Contains failure stories and trust progression protocols. |
cron-patterns.json |
JSON | Defines 5 specific cron jobs for automated repository management. |
skills.md |
Skill Manifest | Lists Tier 1, 2, and 3 skill stacks with installation reasoning. |
privacy-notes.md |
Access Matrix | Details explicit read, write, and network access requirements. |
CodeSmith Advanced Features
- Advanced ACP dispatch patterns for sophisticated multi-agent task delegation.
- Tiered skill management to optimize performance and security within Openclaw Skills.
- Real-world failure recovery strategies integrated directly into the agent's working patterns.
- Incremental trust progression models to safely increase agent autonomy over time.
- Pre-configured cron job templates for automated production monitoring and repository cleanup.
SKILL.md
Loading
Related Openclaw Skills
Solopreneur Assistant
An AI-powered Chief of Staff designed to handle operational overhead, revenue tracking, and strategic decision-making for solo business owners.
Security Hardening
An automated security audit and workspace hardening suite designed to protect AI agents from credential leaks, PII exposure, and prompt injection attacks.
Project Tracker
An automated system for tracking multiple projects with status dashboards, milestone monitoring, and intelligent stalled-project detection.
Memory OS (Persistent AI Agent Memory)
A complete persistent memory system that eliminates context amnesia by giving AI agents long-term recall, identity files, and daily logs.
Streme Token Launcher
A powerful tool for launching SuperTokens on the Base network with integrated liquidity, staking rewards, and vesting vaults.
Clawshi
A real-time prediction market intelligence platform and AI agent competition arena for BTC price movements.
Featured*
