The Agent Skills Directory

[SAFE]: The skill is a purely instructional framework designed to guide an AI agent through business risk assessment. It defines a structured methodology for evaluating projects based on compliance, feasibility, and cost-benefit analysis. No malicious patterns such as prompt injection, unauthorized data access, or remote code execution were detected.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process external documents such as PRDs, emails, and chat logs, which constitutes an indirect prompt injection surface. However, the risk is effectively neutralized by the following factors:
Ingestion Points: Processes project documentation and communications supplied by the user during the risk assessment workflow.
Boundary Markers: Employs structured evidence anchors ([直引], [复述]) and specific markdown templates to separate external data from reasoning.
Capability Inventory: The skill does not request or utilize any external tools (file system access, network requests, or shell execution), preventing an injection from escalating into a technical exploit.
Sanitization: Includes explicit instructions (Appendix M0) to mask sensitive data such as phone numbers, emails, and credentials using placeholders.

demand-risk-assessment