after-effects-cep-panel
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill architecture utilizes a Python-based bridge (PyShiftBridge/PyShiftAE) to perform backend operations. This involves inter-process communication and the execution of Python logic triggered from the JavaScript frontend.
- [DYNAMIC_EXECUTION]: The skill employs the standard Adobe
csInterface.evalScript()method to dynamically execute code strings within the ExtendScript host environment, which is the required method for CEP extension communication. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external data sources such as CSV and JSON files for automation tasks. • Ingestion points: External files (e.g.,
csvFile,videoJsonFile) and project metadata are ingested into the extension environment. • Boundary markers: The architecture usesJSON.stringifyto delimit data during communication between the client and host. • Capability inventory: The skill can execute Python logic and ExtendScript commands within the Adobe After Effects application. • Sanitization: The implementation examples include robust validation and coercion functions (e.g.,_coerce_int,_coerce_bool,_coerce_spacing) to sanitize data before it is processed by the core logic.
Audit Metadata