The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses dynamic context injection (!command) to execute shell commands during skill load for project discovery. It runs find to locate manifest files (e.g., package.json), test directories, and E2E configurations. It also echoes the CI and GITHUB_ACTIONS environment variables. These operations are restricted to project metadata and are consistent with the skill's stated purpose as a test runner.
[DATA_EXFILTRATION]: No network operations or external data transfers were identified. The access to environment variables is limited to non-sensitive identifiers.
[PROMPT_INJECTION]: No attempts to override agent behavior or bypass safety guidelines were found.
[DATA_EXPOSURE]: The skill ingests untrusted data in the form of local file names and environment flags via shell commands. This creates a surface for indirect prompt injection if file names were maliciously crafted, but the skill implements no high-risk capabilities (like network exfiltration) that would make this exploitable.

test-full