aistudio-apps-succession-management
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted metadata from local and remote discovery processes.
- Ingestion points: Data provided by the aistudio tool regarding existing workflows and apps (SKILL.md, references/aistudio-handoff.md).
- Boundary markers: Decisions are delimited by business-facing proposals and user approval gates before any file operations (references/workflow.md).
- Capability inventory: Orchestrates file creation and modification through the aistudio platform based on discovery results.
- Sanitization: Enforces canonical display names and naming conventions for generated artifacts.
- [DYNAMIC_EXECUTION]: The skill generates application code containing ora.Invoke references based on predefined action maps. This is a core function of the app-building purpose and is restricted to newly created current-flow artifacts.
- [SAFE]: Implements a robust local-first discovery rule that prioritizes local project files over remote environments, protecting against remote resource injection.
- [SAFE]: Prohibits any modification of existing artifacts, reference apps, or the base aistudio tool, ensuring project integrity.
- [SAFE]: Requires explicit user approval for a consolidated build proposal before initiating any write operations.
Audit Metadata