The Agent Skills Directory

[SAFE]: The skill's primary purpose is the generation and conversion of project requirements into structured formats (Markdown and JSON). It does not contain any instructions that attempt to bypass AI safety guidelines or exfiltrate data.
[COMMAND_EXECUTION]: The skill utilizes the Bash tool to execute ralph CLI commands for creating, converting, and checking the status of PRDs. These commands are purpose-built for the skill's functionality and do not involve arbitrary shell execution or privilege escalation.
[INDIRECT_PROMPT_INJECTION]: The skill ingests feature descriptions from users to populate PRDs. While this represents a surface for indirect prompt injection, the capability inventory is restricted to local file operations and project-specific CLI tools, posing a low risk to the system environment. No evidence of malicious interpolation or unsafe sanitization was found.

prd