ae-expressions

Pass

Audited by Gen Agent Trust Hub on Oct 16, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [DYNAMIC_EXECUTION]: The skill provides templates for using ae_set_expression and ae_run_jsx to inject and execute code at runtime. Specifically, ae_run_jsx allows for the execution of arbitrary ExtendScript code to manipulate After Effects project items and properties.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external sources and project metadata.
  • Ingestion points: Loads data from external files via footage("data.csv") and retrieves context from user-defined layer names (e.g., thisComp.layer("Name")) and markers.
  • Boundary markers: No specific delimiters or boundary markers are instructed for the agent to use when interpolating content from these untrusted data sources.
  • Capability inventory: The skill utilizes powerful bridge tools including ae_set_expression, ae_run_jsx, and ae_menu_command (all documented in SKILL.md) which can modify project state or execute scripts.
  • Sanitization: There is no evidence of input validation or sanitization for data retrieved from CSV files or project metadata before it is used in expression logic.
  • [COMMAND_EXECUTION]: The skill uses ae_menu_command to trigger application-level menu actions (such as "Convert Expression to Keyframes") and ae_run_jsx to execute host-side scripts.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 16, 2026, 01:06 AM