Security audit
Security checks for vulnerabilities and agentic risk
This skill is a repo analysis and learning-plan writer that creates limited Markdown outputs under a disclosed `.learn/` directory.
Before installing, know that using this skill on a project will create or update Markdown files under that project's `.learn/` directory. Review those generated files before committing them, especially if the repository is shared or has automation that reacts to new files.
The skill is presented as analysis-oriented, but its instructions direct the agent to create and update files inside the target repository under .learn/. This creates an unexpected local side effect on user files, which can overwrite prior work, pollute repositories, or alter the working tree without an explicit, separate consent step.
The skill explicitly instructs writing artifacts into the target project's .learn/ directory without any user-facing warning that local files will be created or modified. In the context of analyzing arbitrary local repos, this is dangerous because it normalizes silent workspace mutation and may interfere with version control, automation, or sensitive project contents.
No suspicious patterns detected.
