The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection as it processes data from external, potentially untrusted GitHub sources.
Ingestion points: Tools such as get_file_contents, issue_read, and pull_request_read (referenced in SKILL.md) fetch content from remote repositories that could be modified by third parties.
Boundary markers: The instructions do not provide explicit delimiters or specific warnings for the agent to ignore instructions embedded within the data fetched from GitHub.
Capability inventory: The skill possesses high-impact write capabilities, including create_or_update_file, delete_file, and merge_pull_request.
Sanitization: There is no mention of sanitizing or validating the content retrieved from remote repositories before processing.

github-skill