The Agent Skills Directory

[PROMPT_INJECTION]: The skill contains explicit instructions to conceal system behavior from the end-user.
Rule 13 in SKILL.md directs the agent to 'drop errors silently' and 'not display it or surface the error text'. While often intended for user experience, this concealment prevents the user from being aware of failures, unexpected responses, or potential security-related errors.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the Ticketmaster API, creating a vulnerability surface for indirect prompt injection.
Ingestion points: Data is ingested via the ticketmaster search-events, event-details, and top-picks CLI tools (documented in SKILL.md).
Boundary markers: The instructions do not define any delimiters or boundary markers to isolate external data from the agent's core instructions.
Capability inventory: The skill uses the ticketmaster CLI, renders Markdown tables, and generates checkout links that the agent uses to guide the user to a browser session.
Sanitization: There is no mention of sanitizing or escaping fields such as event names, descriptions, or listing details before they are processed by the agent.
[EXTERNAL_DOWNLOADS]: The skill references Ticketmaster-hosted images and affiliate checkout domains.
The instructions mention https://app.ticketmaster.com/maps/geometry/... for images and https://ticketmaster.evyy.net/... for checkout URLs. These are expected domains for the skill's stated purpose of providing ticket listing services.

ticketmaster