The Agent Skills Directory

[SAFE]: The skill defines a structured workflow for requirement analysis and communication. The instructions focus on improving the clarity and precision of the agent's output by identifying missing technical details or ambiguous terminology in user-provided tasks.
[SAFE]: No malicious patterns such as prompt injection, data exfiltration, or obfuscation were detected. References to configuration files like ~/.claude/settings.json and environment variables in the header are descriptive of the agent's operating environment and do not constitute instructions for the agent to access or modify sensitive files.

ask-questions-if-underspecified