hackathon-event-parser

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). This skill's SKILL.md requires fetching and parsing arbitrary public hackathon event pages (e.g., Devpost, DoraHacks, Hackathon.com) from the provided event_url and using that untrusted, user-published content to populate outputs that drive the downstream autonomous pipeline (hackathon-track-analyzer), which clearly allows third-party page text to influence tool use and decisions and therefore could enable indirect prompt injection.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly fetches and parses the user-provided event_url at runtime (e.g., https://globalaihealth.devpost.com) and uses that page content to generate the downstream pipeline prompts/instructions, so remote page content can directly control agent behavior.