fix-agent-qa-ticket
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data from ticket comments and conversation transcripts which could influence agent behavior.
- Ingestion points: Data is ingested in Step 1 (
/conversation-triage-tickets) and Step 2 (/conversations) via API calls. - Boundary markers: The instructions do not define explicit delimiters or instructions to ignore embedded commands within the ingested data.
- Capability inventory: The skill has the ability to create branches, modify platform procedures, update system prompts, and post comments back to the ticket system via the ElevenLabs REST API.
- Sanitization: There is no evidence of data sanitization or validation performed on the feedback or transcript content before it is used to inform the fixing process.
- [COMMAND_EXECUTION]: The skill uses
curlto perform REST API operations against ElevenLabs infrastructure for both data retrieval and platform modification. - [EXTERNAL_DOWNLOADS]: Fetches configuration and conversation data from
api.elevenlabs.io. These operations target official vendor infrastructure.
Audit Metadata