The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the agent to fetch updated documentation from the official Render website (render.com). This ensures the agent uses the most current platform specifications for deployments and health checks, targeting only well-known and trusted vendor domains.
[DATA_EXFILTRATION]: The skill contains specific security guidance that warns against passing sensitive credentials via Docker build arguments, as they can be recovered from the image history. It correctly points users toward secure alternatives like secret mounts or runtime environment variables.
[INDIRECT_PROMPT_INJECTION]: The skill defines a protocol for processing external documentation from Render's official website. It mitigates potential risks by providing clear instructions for the agent to verify the retrieval and read the content accurately, ensuring that external content does not override the agent's core safety or operational parameters.

render-docker