prediction-market-oracle-research
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze data from external prediction markets (event descriptions, resolution rules, and tags). This creates a surface for indirect prompt injection if the external sources contain malicious instructions.
- Ingestion points: The 'Research Workflow' steps 2 and 3 involve identifying markets and recording source-grounded probabilities.
- Boundary markers: The skill does not explicitly define delimiters for external content, though it emphasizes source-grounded analysis.
- Capability inventory: The skill does not define or use any tools for file writing, network operations, or command execution.
- Sanitization: No sanitization of ingested market data is specified.
- [SAFE]: No malicious code, obfuscation, or unauthorized access patterns were found. The skill serves as an informational guideline for human-in-the-loop research.
Audit Metadata