The Agent Skills Directory

[SAFE]: No malicious patterns, obfuscation, or hardcoded credentials were found. The skill's behavior is consistent with its stated purpose of verifying code quality.
[COMMAND_EXECUTION]: The skill uses the Bash tool to execute standard development commands such as npm run typecheck, npm run lint, npm test, and npm run build. These operations are local and expected for a build/verification utility.
[PROMPT_INJECTION]: The skill reads package.json and executes the scripts defined within it. This introduces a surface for indirect prompt injection if the project's configuration files are maliciously modified by a third party, as the agent will execute whatever commands are present in the 'scripts' section.
Ingestion points: package.json is read via cat (SKILL.md).
Boundary markers: None present to distinguish between trusted and untrusted script content.
Capability inventory: Access to Bash allows for arbitrary command execution (SKILL.md).
Sanitization: No sanitization or validation of script content is performed before execution.

verify