appbuilder-workfront
Pass
Audited by Gen Agent Trust Hub on Aug 5, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches official developer tools and templates from Adobe's registries, including
@adobe/aio-cli,@adobe/workfront-ui-ext-tpl, and@adobe/uix-guest. - [COMMAND_EXECUTION]: Provides a detailed reference for the
aioCLI to manage the application lifecycle, including authentication, project initialization, local development server execution, and cloud deployment. - [COMMAND_EXECUTION]: Includes instructions for modifying local development environment settings, such as trusting self-signed certificates and disabling the Chrome
local-network-access-checkflag to allow communication between the browser and a local development server. - [PROMPT_INJECTION]: The skill outlines a data flow for processing external information from Workfront, which presents an attack surface for indirect prompt injection.
- Ingestion points: Data enters the system context via
sharedContextin the front-end SPA and theparamsobject in serverless Runtime actions (workfront-actions/assets/action-boilerplate.js). - Boundary markers: The provided code templates do not implement specific delimiters or 'ignore' instructions for data ingested from the Workfront environment.
- Capability inventory: The skill includes capabilities for network requests via
fetchto the Workfront Public API and Planning APIs, as well as file-based configuration management. - Sanitization: Basic validation is included to ensure parameters like the IMS organization ID are present and properly formatted before being used in API headers.
Audit Metadata