The Agent Skills Directory

[SAFE]: The skill identifies and uses project-specific documentation and global configuration files to assist in PRD generation. No malicious behavior such as data exfiltration or unauthorized system access was found.- [PROMPT_INJECTION]: The skill processes conversation context to synthesize the PRD, which is a common pattern for documentation agents. While this creates an indirect prompt injection surface, it is limited by the skill's specific task focus.
Ingestion points: Conversation context and repository files (CLAUDE.md).
Boundary markers: Absent.
Capability inventory: Reading local files and writing to the project issue tracker.
Sanitization: Absent.

to-prd