The Agent Skills Directory

[SAFE]: The skill performs standard product management tasks using a multi-step interactive process. It reads local requirement documents and generates structured user stories based on provided templates. No signs of prompt injection, data exfiltration, or unauthorized command execution were found.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data from PRD and functional design documents (docs/02-方案设计/*.md). While this presents a theoretical attack surface common to all document-processing tools, the skill mitigates risk through mandatory user confirmation steps and strict output templating. No malicious exploitation patterns are present.
[COMMAND_EXECUTION]: The use of the Bash tool is restricted to benign file existence checks and status reporting within the local environment. No dangerous commands or privilege escalation attempts were identified.

pm-user-story