olares-router
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [NO_CODE]: The skill folder contains only markdown documentation and reference files. No executable scripts, binaries, or configuration files that execute code were found within the skill itself.
- [INDIRECT_PROMPT_INJECTION]: The skill describes operations that ingest external, untrusted data which could lead to indirect prompt injection.
- Ingestion points: Scraping URLs (olares-router-call-text.md), processing audio files (olares-router-call-audio.md), and performing OCR on documents (olares-router-calling.md).
- Boundary markers: The documentation refers to platform-level identity injection but does not specify delimiters for data interpolation in prompts.
- Capability inventory: The instructions guide the agent to use olares-cli for network communication and file management.
- Sanitization: Not explicitly addressed in the provided instructions.
Audit Metadata