The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the agent to install the google-play-pp-cli tool using npx from the NPM registry and go install from the vendor's GitHub repository (github.com/mvanhorn/printing-press-library/library/marketing/google-play/cmd/google-play-pp-cli).\n- [COMMAND_EXECUTION]: The skill executes the google-play-pp-cli binary to perform various automated scraping and data analysis tasks on Google Play Store listings.\n- [DATA_EXFILTRATION]: The CLI tool includes a --deliver webhook:<url> feature that allows the agent to transmit scraped results to arbitrary external network endpoints.\n- [DATA_EXFILTRATION]: The CLI tool includes a --deliver file:<path> feature that allows the agent to save results to arbitrary local file paths, which could be used to overwrite sensitive files.\n- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it ingests untrusted data from the public Google Play Store (reviews and app descriptions) that could contain malicious instructions.\n
Ingestion points: Commands such as reviews, app, and watch-listing fetch content directly from the external Play Store (SKILL.md).\n
Boundary markers: The skill does not provide clear delimiters or instructions to ignore embedded commands within the processed data.\n
Capability inventory: The skill uses bash to execute a CLI tool with built-in network exfiltration (--deliver webhook) and file system write (--deliver file) capabilities.\n
Sanitization: There is no mention of sanitizing or validating the scraped external content before it is passed to the agent's context.

pp-google-play