The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes user-generated content (Google Play reviews) which is an untrusted external source. An attacker could craft a review containing malicious instructions that might be interpreted by the AI agent when processing or summarizing the feedback.
Ingestion points: Review data is ingested via gplay reviews list and gplay reviews get in SKILL.md.
Boundary markers: No specific delimiters or instructions to ignore embedded commands are present in the processing logic.
Capability inventory: The skill includes capabilities to write back (gplay reviews reply) and send data to an external email address (mail) as seen in SKILL.md.
Sanitization: No sanitization or filtering of the review text is performed before it is passed to other tools or scripts.
[PERSISTENCE]: The skill provides documentation and scripts for establishing scheduled execution.
Evidence: The SKILL.md file contains a 'Scheduled Review Check' section suggesting the addition of a monitoring script to the system's crontab to ensure daily execution.

gplay-review-management