architect-schedule-launch
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of external content into the agent's environment via knowledge base updates.
- Ingestion points: Data enters the agent's context through KB document creation or rescraping at
POST /v1/convai/knowledge-base/textandPOST /v1/convai/knowledge-base/urlas described in step 4. - Boundary markers: The skill does not provide specific instructions to use delimiters or ignore embedded instructions when processing external text or URL content.
- Capability inventory: The skill uses
curlto interact with the API for reading/writing agent configurations, branch management, and knowledge base attachments. - Sanitization: No explicit sanitization, validation, or filtering of the external knowledge base content is mentioned.
- [COMMAND_EXECUTION]: The skill utilizes
curlshell commands to perform operations against the ElevenLabs API (api.elevenlabs.io). These commands are used to manage agent branches, modify configurations, and update knowledge base documents.
Audit Metadata