The Agent Skills Directory

[CREDENTIALS_UNSAFE]: The skill instructs the agent to read sensitive credentials (AWS_ACCESS_KEY_ID, AWS_SECRET_ACCESS_KEY) from a hardcoded absolute file path: /Users/jonathanburger/remotion/packages/remotion-media/.env. This exposes a specific user's local directory structure and poses a configuration risk for other users.
[COMMAND_EXECUTION]: The skill executes shell commands including git worktree, curl, and bun. It specifically uses bun -e to execute a dynamically generated JavaScript snippet to handle the upload logic.
[DATA_EXFILTRATION]: The skill reads local files and uploads them to a remote Cloudflare R2 bucket (parser-media) to be hosted at the vendor's domain (remotion.media). While this is a documented feature of the skill, it automates the transfer of local data to external storage.
[PROMPT_INJECTION]: The skill processes local data which presents an indirect prompt injection surface.
Ingestion points: The agent reads local files specified by the <local-file> placeholder (SKILL.md).
Boundary markers: None; the skill does not include delimiters or instructions to ignore potential commands embedded within the media assets.
Capability inventory: The skill has the capability to read files, access environment variables, and perform network uploads.
Sanitization: No validation or sanitization is performed on the content of the files before they are processed and uploaded.

upload-r2