The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill installs @vosjs/cli and @vosso/cli via npm, which are resources provided by the skill author for the core functionality of the tool.\n- [EXTERNAL_DOWNLOADS]: It downloads the Chromium browser using npx playwright install chromium and fetches web dependencies such as three and mediabunny from the esm.sh content delivery network during the rendering phase.\n- [COMMAND_EXECUTION]: The instructions require the agent to execute various shell commands including the vos CLI for recording and planning, and ffmpeg for post-processing and re-encoding video assets.\n- [PROMPT_INJECTION]: The skill is subject to indirect prompt injection risks because it directs the agent to interact with and process data from untrusted external websites.\n
Ingestion points: Target web pages explored via fetch or Playwright tools (SKILL.md, Step 1).\n
Boundary markers: None identified; the instructions do not specify the use of delimiters or isolation techniques for external web data.\n
Capability inventory: The agent has the ability to execute shell commands (vos, ffmpeg, npm, npx), perform network operations, and write to the local file system (take directory).\n
Sanitization: There are no instructions for sanitizing or validating content fetched from external URLs before the agent processes it to identify selectors and plan actions.

product-video