Security audit
Security checks for vulnerabilities and agentic risk
This is a markdown-only skill that presents codebase health as a weather report and does not include code execution, credentials, persistence, or hidden data movement.
This skill appears safe to install as a lightweight markdown instruction set. Expect it to read or reason about repository health signals when invoked; review its output before acting on any suggested risk areas, especially before deploys or planning decisions.
Skill instructs the agent to omit warnings, disclaimers, or ethical commentary. Stripping safety caveats hides risk from the user and is a common jailbreak preamble.
Could destroy everything if it breaks. No warning.
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.
# Code Weather
> "You wouldn't go outside without checking the weather. Why would you start coding without checking the forecast?"
## What It Does
The 'When to Invoke' section says to use the skill 'Every morning,' during standup, before deploys, and weekly, but it does not define a specific trigger phrase, command, or exclusion condition. This makes activation scope ambiguous and increases the chance of unintended invocation from general discussion about coding routines or team ceremonies.
No suspicious patterns detected.