The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the agent to download and install external libraries from public registries like PyPI and npm using standard package managers.
Evidence: Bash examples at the end of SKILL.md suggest using pip install package and npm install package.
[COMMAND_EXECUTION]: The skill provides explicit shell commands for the agent to execute in the host environment, including package managers and security scanners.
Evidence: The skill body references execution of pip-audit, npm audit, snyk test, grype, and cargo audit.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and act upon data from external sources (CVE databases, audit tool outputs, and package advisory pages), creating a potential for indirect prompt injection if those sources contain malicious instructions.
Ingestion points: Output from tools like pip-audit, snyk, and grype, as well as package advisory web pages referenced in SKILL.md.
Boundary markers: None identified. The instructions do not define specific delimiters or warnings for the agent to ignore instructions embedded in the audit results.
Capability inventory: The skill has the capability to install packages, audit the filesystem, and generate detailed reports for the user.
Sanitization: No sanitization or validation logic is specified for the data retrieved from external security tools or advisory sources.

vulnerable-components