Skip to content

⚡ SOTA agent skills for OpenClaw — observability, security, code quality, incident response, and more. Built by Anvil AI.

License

Notifications You must be signed in to change notification settings

cacheforge-ai/cacheforge-skills

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

33 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

⚡ Agentic Skills

License: MIT Skills ClawHub OpenClaw Discord Live Lab

Production-grade agent skills for OpenClaw and the ClawHub marketplace.

Built by Anvil AI — R&D enterprise for agentic infrastructure.


Battle-tested on a 5-agent autonomous swarm running 24/7. If it ships here, it survived production first.


Skills

🔍 Code Quality & Security

Skill Description Version Install
vibe-check Code quality auditor for risky AI-generated patterns. Catches what linters miss. 0.1.2 clawhub install vibe-check
dep-audit Dependency vulnerability scanner across npm, pip, Cargo, and Go. 0.1.4 clawhub install dep-audit
rug-checker Solana token risk analysis with visual scoring. Trust, but verify. 0.1.4 clawhub install rug-checker

📡 Observability & Infrastructure

Skill Description Version Install
prom-query Prometheus metrics query and alert interpreter. 1.0.2 clawhub install prom-query
kube-medic Kubernetes cluster triage and diagnostics. First responder for your pods. 1.0.2 clawhub install kube-medic
log-dive Unified log investigation (Loki, Elasticsearch, CloudWatch). 0.1.2 clawhub install log-dive
swarm-self-heal Swarm watchdog for gateway/channel/lane liveness with bounded recovery receipts. 0.1.1 clawhub install swarm-self-heal
tf-plan-review Terraform plan risk assessment before apply. 0.1.2 clawhub install tf-plan-review

🚨 Incident Response

Skill Description Version Install
pager-triage PagerDuty incident triage and on-call workflow support. 0.1.2 clawhub install pager-triage

🧠 Productivity

Skill Description Version Install
feed-diet Information-diet analysis across HN and RSS/OPML. Cut the noise. 0.1.2 clawhub install feed-diet
meeting-autopilot Transcript-to-action pipeline — decisions, owners, follow-ups. No more "what did we decide?" 0.1.2 clawhub install meeting-autopilot

🛠️ Core Engineering

Skill Description Version Install
agentic-devops Docker, process, log, and health operations toolkit. 1.0.1 clawhub install agentic-devops
context-engineer Context-window analysis and optimization workflows. 1.0.1 clawhub install context-engineer
continuity-kernel Fail-open continuity kernel with deterministic runtime contracts and shadow eval receipts. 0.2.0 clawhub install continuity-kernel

Quick Install

clawhub install agentic-devops
clawhub install context-engineer
clawhub install continuity-kernel
clawhub install log-dive
clawhub install dep-audit

Adding a New Skill

New skills must be categorized into an existing section above (or propose a new category with justification). Every skill needs:

  • A folder under skills/<skill-name>/
  • A SKILL.md with frontmatter (name, description, license, metadata)
  • A row in the appropriate category table in this README
  • Version tracking and a CHANGELOG entry

Follow the existing format exactly — folder layout, metadata schema, table style.

Links

License

MIT

About

⚡ SOTA agent skills for OpenClaw — observability, security, code quality, incident response, and more. Built by Anvil AI.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Sponsor this project

Packages

 
 
 

Contributors