architect-explain-test-runs
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill connects to the official vendor host
https://api.elevenlabs.io. This is a legitimate vendor resource for the author 'elevenlabs'. - [SAFE]: Authentication is managed via a placeholder
$API_KEY, which is a standard and safe method for handling secrets in agent skills. - [INDIRECT_PROMPT_INJECTION]: The skill has the following profile regarding indirect prompt injection:
- Ingestion points: The skill ingests test transcripts and rationales via the
/v1/convai/test-invocations/{suite_id}endpoint (SKILL.md). - Boundary markers: No explicit boundary markers or 'ignore' instructions for the data payload are present.
- Capability inventory: The skill is limited to performing network reads from the vendor API; it has no shell execution, file system write, or arbitrary network capabilities.
- Sanitization: No specific sanitization or filtering of the transcript content is mentioned.
- Ingestion points: The skill ingests test transcripts and rationales via the
Audit Metadata