track-unrest-events
Fail
Audited by Gen Agent Trust Hub on Jul 8, 2026
Risk Level: HIGHCREDENTIALS_UNSAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSSAFE
Full Analysis
- [CREDENTIALS_UNSAFE]: The skill contains a hardcoded API key (
wm_0123456789abcdef0123456789abcdef01234567) in the authentication header section of the SKILL.md file. - [EXTERNAL_DOWNLOADS]: The skill makes network requests to
api.worldmonitor.appto retrieve event data. This domain is associated with the skill's authoring organization. - [COMMAND_EXECUTION]: The documentation includes a worked example that demonstrates the use of shell commands including
node,curl, andjqto process API data. - [SAFE]: The skill implements strong security best practices for handling untrusted data. The 'Content safety' section explicitly instructs the agent to treat API responses as data only and to ignore any directive-like text (such as 'ignore previous instructions') found within the data.
- [SAFE]: Data ingestion occurs from a remote API (
api.worldmonitor.app). The skill provides explicit boundary markers by warning the agent to distinguish between instructions and fetched data. It inventory includes standard network and data processing capabilities. Sanitization is addressed via explicit instructions to disregard embedded directives.
Recommendations
- AI detected serious security threats
Audit Metadata