The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands to trigger script execution in Adobe After Effects.
On Windows, it calls AfterFX.exe with the -r flag to run a .jsx file.
On macOS, it uses osascript to send a DoScript command to the Adobe After Effects application.
[DYNAMIC_EXECUTION]: The core functionality of the skill involves the agent dynamically generating ExtendScript (.jsx) files at runtime and executing them via the host environment. The skill provides specific guidance on handling the legacy ES3 JavaScript engine used by After Effects.
[INDIRECT_PROMPT_INJECTION]: The skill establishes a feedback loop where the agent reads the content of ae/result.json after execution. This creates an attack surface where data or error messages generated by the script are ingested back into the agent's context.
Ingestion points: The agent is instructed to read ae/result.json in the work/ directory.
Boundary markers: No specific delimiters or warnings for the agent to ignore instructions within the JSON result are provided.
Capability inventory: The skill has the ability to execute shell commands and perform file operations within the Adobe After Effects environment.
Sanitization: The provided JSX wrapper performs basic sanitization by replacing double quotes in error messages before writing them to the JSON file.

ae-extendscript