web-webauthn-software-authenticator
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill describes processing data received from external web endpoints, which introduces a surface for indirect prompt injection from untrusted Relying Party (RP) servers.
- Ingestion points: The skill ingests 'options' (challenge, rp.id, user.id) from the target server's /register/begin and /auth/begin responses as described in SKILL.md.
- Boundary markers: No specific delimiters or validation checks are defined for the incoming server-controlled data.
- Capability inventory: The skill involves generating cryptographic signatures and performing network requests to interaction with the target application.
- Sanitization: The instructions do not specify sanitization or escaping of the server-controlled data before it is used in the authenticator logic or subsequent responses.
Audit Metadata