The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute various package manager CLI tools, including npm, pip, pip-audit, mvn, gradle, and cargo. These commands are standard for the skill's intended purpose of dependency management and auditing.
[INDIRECT_PROMPT_INJECTION]: The skill processes external project configuration files which serve as an entry point for untrusted data.
Ingestion points: Files such as package.json, requirements.txt, and pyproject.toml are ingested by the agent and the scripts/parse-dependencies.py script.
Boundary markers: The instructions lack explicit boundary markers or directives for the agent to treat natural language content within these files (such as descriptions or comments) as data rather than instructions.
Capability inventory: The skill has the capability to write to the file system (updates) and execute shell commands (audits and installations).
Sanitization: There is no evidence of sanitization or validation of the metadata within dependency files before the agent processes them, which could allow a malicious file to attempt to mislead the agent during an audit.

dependency-management