The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests content from external URLs provided by the user, which serves as a vector for indirect prompt injection attacks.
Ingestion points: The url parameter defined in SKILL.md and the source.url field within the render_video tool arguments.
Boundary markers: There are no explicit delimiters or specific instructions provided to the agent to disregard instructions that might be contained within the article body.
Capability inventory: The skill utilizes the render_video and get_project_status MCP tools, and provides a utility script (examples/run.sh) capable of performing network operations via curl.
Sanitization: No sanitization or filtering of the article content is implemented before it is passed to the video generation workflow.
[COMMAND_EXECUTION]: The skill includes a shell script (examples/run.sh) that executes commands to interact with the Revid API.
The script uses curl to perform HTTP POST and GET requests to https://www.revid.ai/api/public/v3/ for rendering and status tracking.
The script utilizes jq for parsing JSON responses.
These commands are intended for manual testing and smoke testing of the API integration.

revid-article-to-short