The Agent Skills Directory

[SAFE]: The skill performs standard development tasks using natural language instructions. No evidence of credential theft, data exfiltration to unauthorized domains, or privilege escalation was found.\n- [NO_CODE]: The skill consists entirely of markdown-based instructions for the AI agent. It does not include or execute any scripts, binaries, or external code dependencies.\n- [PROMPT_INJECTION]: The skill involves reading content from the user's repository ("Explore the repo to understand the current state of the codebase") and publishing output to an external service ("publish it to the project issue tracker"). While this creates an indirect prompt injection surface where malicious code in the repository could attempt to influence the agent's PRD generation, this is a core requirement for the tool's functionality. No specific exploitable patterns were identified, and the risk is considered low and manageable via standard agent safety protocols.

to-prd