Ops Skills Pack

# OpenClaw Ops Skills Pack > **Production-ready skills for autonomous agent operations** ## Metadata **Name**: Ops Skills Pack **Version**: 1.0.0 **Category**: Operations **Author**: Eric Jie <jxmerich@mail.com> **License**: MIT **Compatibility**: OpenClaw >= 2.0.0 **Repository**: https://github.com/Erich1566/openclaw-ops-skills ## Description Transform your OpenClaw agent from a chatbot into production-ready autonomous infrastructure. Based on 1.4+ billion tokens of real-world testing, this skill pack provides the guardrails, workflows, and best practices needed for autonomous operations that work while you sleep. ## Skills Included (14) ### Core Infrastructure 1. **model-routing** - Optimize model selection for 80% cost reduction 2. **execution-discipline** - Enforce Build→Test→Document→Decide cycle 3. **docs-first** - Mandatory reconnaissance before code changes 4. **scope-control** - Define and enforce operational boundaries 5. **task-autonomy** - Self-expanding task management 6. **progress-logging** - Comprehensive execution logs ### Operations 7. **memory-persistence** - File-based memory system 8. **cron-orchestration** - Scheduled autonomous operations 9. **error-recovery** - Graceful failure handling 10. **security-hardening** - Security best practices ### Advanced 11. **cost-optimization** - Token waste prevention 12. **testing-protocol** - Quality assurance standards 13. **communication** - Structured update templates 14. **integration-guide** - Safe third-party connections ## Key Benefits - **80% Cost Reduction** - Intelligent model routing (Sonnet 4.6 for daily work) - **True Autonomy** - Self-expanding task lists continue work overnight - **Persistent Memory** - Survives session compression - **Complete Visibility** - Morning briefings replace chat history scrolling - **Security Hardened** - Built-in security practices ## Quick Start ```bash # Install skills cp skills/*.md ~/.openclaw/workspace/skills/ # Set up workspace templates cp templates/*.md ~/.openclaw/workspace/ # Configure model routing (see README.md) # Set up cron jobs openclaw cron add --name "overnight-2am" --cron "0 2 * * *" \ --message "Check Todo.md. Continue work. Log progress." # Restart openclaw restart ``` ## Requirements - OpenClaw 2.0+ - 15 minutes for initial setup - Willingness to treat configuration as work ## Documentation - **README.md** - Full documentation - **QUICKSTART.md** - 15-minute setup guide - **FEATURES.md** - Bilingual feature guide - **DESCRIPTION.md** - Project overview ## Installation Order **Week 1**: Core skills (model-routing, execution-discipline, scope-control) **Week 2**: Operational skills (docs-first, task-autonomy, progress-logging) **Week 3**: Advanced skills (memory-persistence, cron-orchestration, error-recovery) ## Support - GitHub: https://github.com/yourusername/openclaw-ops-skills - Issues: Report bugs and request features - Discussions: Ask questions and share experiences

OpenClaw Ops Skills Pack

Production-ready skills for autonomous agent operations

Metadata

Name: Ops Skills Pack Version: 1.0.0 Category: Operations Author: Eric Jie jxmerich@mail.com License: MIT Compatibility: OpenClaw >= 2.0.0 Repository: https://github.com/Erich1566/openclaw-ops-skills

Description

Transform your OpenClaw agent from a chatbot into production-ready autonomous infrastructure. Based on 1.4+ billion tokens of real-world testing, this skill pack provides the guardrails, workflows, and best practices needed for autonomous operations that work while you sleep.

Skills Included (14)

Core Infrastructure

model-routing - Optimize model selection for 80% cost reduction
execution-discipline - Enforce Build→Test→Document→Decide cycle
docs-first - Mandatory reconnaissance before code changes
scope-control - Define and enforce operational boundaries
task-autonomy - Self-expanding task management
progress-logging - Comprehensive execution logs

Operations

memory-persistence - File-based memory system
cron-orchestration - Scheduled autonomous operations
error-recovery - Graceful failure handling
security-hardening - Security best practices

Advanced

cost-optimization - Token waste prevention
testing-protocol - Quality assurance standards
communication - Structured update templates
integration-guide - Safe third-party connections

Key Benefits

80% Cost Reduction - Intelligent model routing (Sonnet 4.6 for daily work)
True Autonomy - Self-expanding task lists continue work overnight
Persistent Memory - Survives session compression
Complete Visibility - Morning briefings replace chat history scrolling
Security Hardened - Built-in security practices

Quick Start

# Install skills
cp skills/*.md ~/.openclaw/workspace/skills/

# Set up workspace templates
cp templates/*.md ~/.openclaw/workspace/

# Configure model routing (see README.md)

# Set up cron jobs
openclaw cron add --name "overnight-2am" --cron "0 2 * * *" \
  --message "Check Todo.md. Continue work. Log progress."

# Restart
openclaw restart

Requirements

OpenClaw 2.0+
15 minutes for initial setup
Willingness to treat configuration as work

Documentation

README.md - Full documentation
QUICKSTART.md - 15-minute setup guide
FEATURES.md - Bilingual feature guide
DESCRIPTION.md - Project overview

Installation Order

Week 1: Core skills (model-routing, execution-discipline, scope-control) Week 2: Operational skills (docs-first, task-autonomy, progress-logging) Week 3: Advanced skills (memory-persistence, cron-orchestration, error-recovery)

Support

GitHub: https://github.com/yourusername/openclaw-ops-skills
Issues: Report bugs and request features
Discussions: Ask questions and share experiences

Ops Skills Pack

Safety Notice

Copy this and send it to your AI assistant to learn