Security audit
Security checks for vulnerabilities and agentic risk
This skill is a code-review workflow helper and does not show evidence of hidden execution, persistence, credential access, or data exfiltration.
Install this if you want the agent to proactively request code reviews during development. Expect it to inspect git diffs and share them with a review subagent; review frequency may be higher than you want unless you tune or override the workflow.
The skill description is broadly phrased to trigger during common development milestones like completing tasks, implementing features, or preparing to merge. Because these conditions are routine and not narrowly scoped to a specific repository, workflow, or explicit user request, the skill may activate more often than intended and steer agent behavior in ways the user did not ask for.
Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.
- Give clear verdict
**DON'T:**
- Say "looks good" without checking
- Mark nitpicks as Critical
- Give feedback on code you didn't review
- Be vague ("improve error handling")
No suspicious patterns detected.