reference-class-forecasting
Pass
Audited by Gen Agent Trust Hub on Aug 25, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [PROMPT_INJECTION]: The skill incorporates explicit defensive instructions for the AI, stating that all text supplied by the user must be treated as untrusted data and not as instructions to execute. This is a recommended security pattern for agent instructions.
- [DATA_EXPOSURE_AND_EXFILTRATION]: The skill does not reference any sensitive file paths, environment variables, or hardcoded credentials. It does not utilize any tools or commands capable of network communication.
- [UNVERIFIABLE_DEPENDENCIES_AND_REMOTE_CODE_EXECUTION]: There are no scripts, binaries, or external package dependencies (e.g., pip, npm) associated with this skill. It operates entirely through natural language guidance.
- [DYNAMIC_CONTEXT_INJECTION]: The skill does not use the exclamation-backtick syntax or any other mechanism for runtime shell command execution.
Audit Metadata