The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute various pz CLI commands to list projects, view feeds, and retrieve paper content.\n- [EXTERNAL_DOWNLOADS]: The documentation provides links to official vendor sources for the pz CLI, including GitHub repositories and the vendor's documentation site.\n- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it ingests and processes content from external research papers and project feeds.\n
Ingestion points: Untrusted content enters the agent's context through the pz paper and pz feed commands (SKILL.md).\n
Boundary markers: No specific delimiters or instructions to disregard embedded content are provided.\n
Capability inventory: The agent can execute multiple platform commands and submit feedback based on processed data.\n
Sanitization: There is no evidence of sanitization or filtering applied to the external data before interpolation.

paperzilla