The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses standard command-line utilities such as grep, git, and package manager audit tools (pnpm, npm, yarn, bun) to perform static analysis and audit project dependencies. It also references jstar detect, a tool associated with the vendor 'j-star-films-studios'.
[EXTERNAL_DOWNLOADS]: The skill uses npm view to fetch metadata from the public npm registry. This is used to verify the existence and popularity of dependencies to mitigate supply chain risks like typosquatting.
[DATA_EXPOSURE]: The skill provides instructions to scan the local filesystem for sensitive files (e.g., .env, .pem) and hardcoded credentials using regex patterns, ensuring these secrets are not tracked in version control.
[INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it reads and processes untrusted external codebases. 1. Ingestion points: Analyzes project files, lockfiles, and git history using grep and git ls-files. 2. Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded prompts in the code being audited. 3. Capability inventory: The agent has access to shell execution (grep, npm, git, jstar) and filesystem read capabilities. 4. Sanitization: There is no explicit sanitization of the content read from files before it is processed by the agent.

security-audit