advanced-app-studio
Pass
Audited by Gen Agent Trust Hub on Apr 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
community-domo-clitool andpython3to automate complex operations such as app structure retrieval, layout updates, and card creation. - [COMMAND_EXECUTION]: In-line Python script generation is employed to manipulate JSON data structures for dashboard layouts and card definitions, executed via heredocs and CLI pipes.
- [PROMPT_INJECTION]: The skill ingests untrusted configuration and schema data from the Domo API, creating a surface for indirect prompt injection where malicious input in an instance could influence agent behavior.
- Ingestion points: Data retrieved from
community-domo-cli app-studio get,layout-get,cards definition, anddatasets schema. - Boundary markers: Absent; the skill does not use delimiters or warnings to isolate ingested data from instructions.
- Capability inventory: Significant modify-access capabilities including
app-studio update,layout-set,cards create, andcreate-view. - Sanitization: None; data is processed using standard JSON parsing without validation of content fields.
Audit Metadata