// the find
BagelHole/DevOps-Security-Agent-Skills
Agent-ready DevOps, security, infrastructure, and compliance knowledge base with 80+ skills across Kubernetes, Terraform, AWS/Azure/GCP, AI platform operations, container hardening, SOC2/ISO27001, and incident response—plus ready-to-run scripts, templates, and playbooks for SRE, platform, and security teams.
A library of 160+ 'Agent Skills' — structured Markdown + scripts covering Kubernetes, Terraform, the big three clouds, compliance frameworks (SOC2/HIPAA/PCI), and AI/LLMOps ops — meant to be dropped into Claude Code, Cursor, or Codex so the agent has copy-pasteable configs and playbooks on hand. Aimed at platform, SRE, and security engineers who want their coding agent to stop hallucinating Terraform and start citing real HCL.
Every skill follows the same SKILL.md + scripts/ + references/ + assets/ layout, so an agent can index and selectively load them without parsing 160 different structures. The content is genuinely runnable — real YAML/HCL/bash, not just links to docs, and installs in one line via `npx skills add`. It bundles topics you don't usually see under one roof, like LLM cost optimization and agent evals sitting next to classic Terraform/K8s material.
Breadth this wide means depth is necessarily shallow — one SKILL.md isn't a substitute for the Vault policy docs or the actual FedRAMP control catalog, and there's no versioning or last-verified date on any skill, which matters a lot for cloud APIs and CVE tooling that shift constantly. Letting an agent auto-load and execute hardening/audit scripts from a third-party repo with no CI, no tests, and no signed provenance is a real supply-chain risk — mildly ironic given one of the included skills is literally about supply-chain attack response. The README's tone and the uniform table structure across 160+ skills strongly suggest bulk LLM generation rather than field-tested runbooks, so quality is unverified until you actually read the one you need.