The Agent Skills Directory

[SAFE]: The skill follows security best practices by requiring explicit user confirmation of all booking details (hotel, room, price, and guest information) before calling the create_booking tool to initiate a transaction.
[DATA_EXFILTRATION]: The skill collects PII (full name, phone number, and email) for the purpose of completing hotel reservations. However, it implements privacy safeguards by mandating that this data must be explicitly typed by the user rather than inferred from account profiles, and it restricts data transmission to the authorized pricewin MCP tool environment.
[COMMAND_EXECUTION]: Analysis of the skill instructions and references shows no evidence of unauthorized shell command execution, privilege escalation, or persistence mechanisms. All operations are performed through scoped MCP tool calls.
[REMOTE_CODE_EXECUTION]: No remote code execution patterns, such as piping network downloads to interpreters or dynamic code generation from untrusted sources, were identified.
[PROMPT_INJECTION]: The skill contains clear instructional boundaries and safety requirements that mitigate potential injection risks. It specifically instructs the agent to ignore profile-based identity data in favor of manual user input for sensitive fields.

pricewin-booking-assistant