Is 0xameer/cloud-devops safe?

https://github.com/openclaw/skills/tree/main/skills/0xameer/cloud-devops

97
SAFE

This skill is a personal resume/CV for a DevOps engineer, not a functional agent skill. It contains no executable code, no prompt injection techniques, no data exfiltration mechanisms, and no malicious behavior. All monitoring signals (network, filesystem, process execution, canary integrity) are clean. The only concern is that it consumes system prompt context with non-functional content.

Category Scores

Prompt Injection 95/100 · 30%
Data Exfiltration 100/100 · 25%
Code Execution 100/100 · 20%
Clone Behavior 95/100 · 10%
Canary Integrity 100/100 · 10%
Behavioral Reasoning 85/100 · 5%

Findings (3)

INFO Skill is a resume, not functional agent instructions -15 ▶

The SKILL.md contains a personal CV/resume for 'Sameer Gupta' listing DevOps skills, competencies, and languages. It provides no agent instructions, tool definitions, or behavioral directives. This is a misuse of the skill format but poses no security risk.

LOW Implicit persona context injection via resume content -5 ▶

When injected into an agent's system prompt, this resume content may subtly influence the agent to adopt a DevOps-expert persona or weight its responses toward the listed technologies (Kubernetes, Terraform, NixOS, etc.). This is not an explicit prompt injection but could mildly bias agent behavior. No override instructions or adversarial techniques are present.

INFO Standard SDK cache files created during installation -5 ▶

Installation created JIT compilation cache files in /tmp/jiti/ which are standard ClawHub SDK runtime artifacts. These are not skill-initiated and represent normal platform behavior during the sparse checkout and install process.