The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the user to install a dependency (writing-skills) from an untrusted third-party GitHub repository: https://github.com/obra/superpowers.
[COMMAND_EXECUTION]: The skill executes bundled shell and Python scripts (scripts/validate.sh and scripts/quick_validate.py) to automate structural checks on skill files.
[PROMPT_INJECTION]: The skill has an indirect prompt injection surface because it reads and analyzes content from other potentially untrusted skills.
Ingestion points: Reads SKILL.md and script files from the target skills being audited.
Boundary markers: Absent. The auditing instructions do not specify the use of delimiters or warnings to ignore instructions found within the audited content.
Capability inventory: The skill can execute local bash/python scripts and invoke other skills (writing-skills).
Sanitization: Absent. The validation logic checks metadata format and length but does not sanitize the instructional content of audited files.

skill-reviewer