The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the user to execute npx -y @helios-project/cli@latest, which downloads and runs code directly from the NPM registry from an unverified organization.
[COMMAND_EXECUTION]: The workflow relies on executing shell commands (npx, node, ts-node) that perform file system operations (writing output.mp4) and network requests to fetch compositions.
[INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted external content that could influence agent behavior.
Ingestion points: The compositionUrl (which can be a remote URL or local file) and the inputProps object used for dynamic content injection in SKILL.md.
Boundary markers: None present; the skill does not define delimiters or safety warnings for the content being rendered.
Capability inventory: The skill has the capability to write files to disk (outputPath) and execute shell commands (npx, node).
Sanitization: There is no evidence of sanitization or validation of the input properties or the composition content before rendering.

render-video