The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses subprocess.run to execute local git commands for repository management tasks such as fetching, pulling, and checking commit histories.
[EXTERNAL_DOWNLOADS]: The skill retrieves data from github.com (for marketplace metadata and commit SHAs) and skills.sh (for trending skill lists) using urllib.request.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it processes untrusted data from external sources that could contain malicious instructions designed to influence the agent's behavior.
Ingestion points: scripts/check_updates.py (fetches remote marketplace.json and commit info) and scripts/recommend_skills.py (scrapes HTML from skills.sh).
Boundary markers: No explicit delimiters or instructions are provided to the agent to disregard potential instructions embedded in the external data.
Capability inventory: The skill possesses the capability to execute shell commands (via git) and suggest installation commands to the user or agent.
Sanitization: While the skill uses HTMLParser and regular expressions to extract relevant fields, it does not implement specific sanitization to detect or prevent prompt injection attacks.

skills-updater