The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes user reviews from the Google Play Store via commands such as gpc reviews list and gpc reviews analyze. These reviews represent untrusted external data that could contain malicious instructions intended to influence the agent's actions or summaries. 1. Ingestion points: Review content is fetched and analyzed as described in SKILL.md and references/review-management.md. 2. Boundary markers: The instructions do not specify the use of delimiters or clear separation between review data and agent instructions. 3. Capability inventory: The agent has the capability to execute gpc commands to halt rollouts, reply to reviews, and trigger webhooks. 4. Sanitization: There is no mention of sanitizing or escaping the review text before it is processed by the agent.
[COMMAND_EXECUTION]: The skill relies on executing the gpc CLI tool for its primary functions. It also includes a detection script scripts/detect_gpc.mjs that uses execSync to verify the environment, check authentication status, and read configuration, which is a standard procedure for tools of this nature.
[EXTERNAL_DOWNLOADS]: The documentation provides instructions to install the @gpc-cli/cli package from the NPM registry. This is a standard installation of a vendor-provided tool from a well-known service.

gpc-vitals-monitoring