voice-extractor
Pass
Audited by Gen Agent Trust Hub on Apr 4, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill is designed to ingest and process user-provided writing samples to extract stylistic patterns. This establishes an indirect prompt injection surface as the agent processes untrusted text. However, the risk is negligible because the skill does not use tools for network access, file system modification, or command execution.\n
- Ingestion points: Writing samples are requested as primary inputs in
SKILL.mdandSKILL-OC.md.\n - Boundary markers: The instructions do not define specific delimiters (e.g., XML tags) to separate user data from the system instructions.\n
- Capability inventory: No dangerous tools or platform capabilities are utilized across any of the scripts.\n
- Sanitization: No custom sanitization is performed, relying instead on the model's inherent safety guardrails.\n- [EXTERNAL_DOWNLOADS]: The skill contains informational references to the author's personal site (
brianrwagner.com) and a digital product page ongumroad.com. These are well-known services and are used neutrally for attribution and as a source for the included markdown template, posing no automated execution risk.
Audit Metadata