The Agent Skills Directory

[SAFE]: The skill consists entirely of documentation (Markdown) and static configuration (YAML). There are no executable scripts, subprocess calls, or active network operations.
[DATA_EXPOSURE]: The documentation mentions sensitive file locations and secret stores, specifically .env files and GitHub secrets, in the context of describing how a setup utility (/wizard) functions. These are legitimate references to standard developer practices for secret management and do not involve unauthorized access or exfiltration.
[METADATA_POISONING]: The skill includes external review metadata (REVIEW.md) that references tool safety scores (Snyk, Socket). These are informational records and do not contain deceptive instructions or injection vectors.

ask-matt