The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it is designed to read and modify existing documentation files.
Ingestion points: The skill interacts with existing files in the docs/prds/ directory when a user asks to "regenerate existing PDR".
Boundary markers: There are no defined delimiters or instructions to ignore embedded commands within the files being read, which could lead the agent to follow instructions hidden in those documents.
Capability inventory: The skill is authorized to use the Bash, Write, and Edit tools, providing a high-impact execution path if malicious instructions are ingested.
Sanitization: No sanitization, validation, or filtering of the content within the processed files is specified in the instructions.

PRD Generator