architect-edit-existing-tool
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process configuration data from external API endpoints, creating a potential surface for indirect prompt injection.
- Ingestion points: The skill retrieves tool and agent configurations via GET /v1/convai/tools and GET /v1/convai/agents as described in the SKILL.md instructions.
- Boundary markers: There are no explicit instructions or delimiters provided to ensure the agent ignores natural language instructions that might be embedded within the tool's description or metadata fields during the read-modify-patch cycle.
- Capability inventory: The skill utilizes network capabilities (PATCH requests) to modify agent configurations and tool properties based on the processed data.
- Sanitization: No sanitization or validation logic is specified for the content of the retrieved configurations before they are used to form subsequent API requests.
Audit Metadata