metaculus
Pass
Audited by Gen Agent Trust Hub on Feb 18, 2026
Risk Level: SAFE
Full Analysis
- Prompt Injection (SAFE): No instructions found that attempt to override agent behavior or bypass safety filters. Instructional language is standard and descriptive.
- Data Exposure & Exfiltration (SAFE): The documentation correctly identifies that API tokens should be stored in environment variables rather than hardcoded. The provided
curlexample targets the official Metaculus domain. - Obfuscation (SAFE): No encoded strings, zero-width characters, or homoglyphs detected in the documentation.
- Unverifiable Dependencies & Remote Code Execution (SAFE): No external package installations or remote script executions (e.g., pipe to bash) are requested.
- Privilege Escalation (SAFE): No commands requiring administrative or root privileges are present.
- Persistence Mechanisms (SAFE): No modifications to shell profiles, cron jobs, or startup services are suggested.
- Metadata Poisoning (SAFE): Metadata fields and descriptive text are consistent with the stated purpose of the skill.
- Indirect Prompt Injection (SAFE): While the skill involves retrieving external data from the Metaculus API, it is for the primary purpose of forecasting and data analysis. No specific vulnerability surfaces are exposed in the documentation.
- Time-Delayed / Conditional Attacks (SAFE): No logic gating behavior based on time or environment conditions was found.
- Dynamic Execution (SAFE): The documentation mentions Python functions for CDF generation but does not include any runtime compilation or unsafe deserialization patterns.
Audit Metadata