The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes data such as event names, artist identities, and booking details that could potentially contain malicious instructions if sourced from untrusted external records.
Ingestion points: Data enters the agent's context through event details and traveller information processed by the ticketfairy CLI and the "Fai" assistant (referenced in SKILL.md).
Boundary markers: The skill explicitly instructs the agent to "show what you will create, change, send, refund or cancel and wait for their yes" and to "Keep actions within the organiser's approval."
Capability inventory: The skill utilizes the ticketfairy CLI for account management and event data operations, and an MCP server for guide retrieval.
Sanitization: The skill relies on manual verification and explicit user approval before any state-changing operations are performed, effectively mitigating automated execution of injected instructions.
[EXTERNAL_DOWNLOADS]: The skill references the installation of the ticketfairy package from the NPM registry and connection to the official Ticket Fairy MCP server (ticketfairy.com/docs/mcp). These are official resources provided by the skill's vendor.
[COMMAND_EXECUTION]: The skill provides instructions for using the ticketfairy CLI to perform administrative tasks such as brand selection and event data retrieval. These commands are part of the intended functionality of the official vendor tool.

book-hotels-and-travel-for-the-team