The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses shell commands for environment initialization and session management.
Evidence: Use of uuidgen, cat, and mkdir to manage a persistent client identifier in ~/.config/nemovideo/client_id.
Evidence: Use of curl for authenticated API interactions, including token acquisition and media upload.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted natural language input and external media URLs, creating a standard injection surface.
Ingestion points: Natural language edit descriptions in chat and external media URLs/file paths provided for upload in SKILL.md (§3.2).
Boundary markers: None identified in the prompt interpolation logic.
Capability inventory: Network communication via curl, file read capabilities via curl -F (upload), and state management in the local file system.
Sanitization: No explicit sanitization or validation of the content of the natural language messages or file paths is described.
[DATA_EXFILTRATION]: The skill performs network operations to external domains to support its core functionality.
Evidence: Authenticated requests to https://mega-api-prod.nemovideo.ai and interaction with https://nemovideo.com for session claiming and status monitoring.

ai-video-editor