rate-limiting-apis
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to use Read and Grep on access logs to analyze traffic patterns. Since access logs are populated by external user requests, they represent an untrusted data source that could contain adversarial content or instructions intended to influence the agent during analysis.
- [EXTERNAL_DOWNLOADS]: References external documentation from the IETF and the
rate-limiter-flexiblelibrary on GitHub. These are well-known, legitimate resources for software development. - [COMMAND_EXECUTION]: The skill requests access to a restricted Bash tool (
api:ratelimit-*) for performing implementation tasks. This is a controlled capability but allows for command execution on the host system.
Audit Metadata