run-giveaway
Pass
Audited by Gen Agent Trust Hub on Apr 27, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns or security vulnerabilities were identified. The skill is configured for API-only execution, limiting its access to the official vendor domain.
- [INDIRECT_PROMPT_INJECTION]: The skill handles untrusted data from X (Twitter) such as tweet text and profile bios.
- Ingestion points: External data enters through API responses when fetching entrants and tweet details in SKILL.md.
- Boundary markers: The skill includes a dedicated Security section explicitly instructing the agent to treat tweet content as untrusted and to never execute instructions found within it.
- Capability inventory: The skill is restricted to network operations with no capabilities for file system access, shell execution, or local scripts.
- Sanitization: Security instructions provide a clear barrier to prevent the agent from obeying instructions embedded in the processed data.
Audit Metadata