hackathon-event-parser
Pass
Audited by Gen Agent Trust Hub on Mar 12, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- [PROMPT_INJECTION]: The skill is designed to fetch and parse content from external URLs provided in the
event_urlparameter. This exposes the agent to indirect prompt injection, where malicious instructions embedded in a hackathon page (e.g., within track descriptions, judging criteria, or sponsor briefs) could attempt to hijack the agent's logic or influence the behavior of the autonomous pipeline it triggers.\n - Ingestion points: External data enters through the
event_urlinput inSKILL.md.\n - Boundary markers: Absent. There are no instructions or delimiters provided to help the agent distinguish between parsing rules and the data retrieved from the external source.\n
- Capability inventory: No risky system capabilities (e.g., file writing, network requests, shell access) are defined within this specific skill file, though it acts as a trigger for a downstream autonomous pipeline.\n
- Sanitization: Absent. The skill extracts raw content directly from the event page without filtering or escaping potentially malicious instructions.\n- [NO_CODE]: The skill contains only natural language instructions and YAML metadata; no executable scripts, binaries, or source code files are included.
Audit Metadata