The Agent Skills Directory

[COMMAND_EXECUTION]: The skill's primary function is to generate and execute shell commands using the steel browser utility to perform web interactions, manage session states, and evaluate JavaScript within a browser context.
[EXTERNAL_DOWNLOADS]: Installation instructions provided in the README guide users to fetch the skill from the author's official GitHub repository (steel-dev/cli), which is a legitimate resource for the skill's components.
[COMMAND_EXECUTION]: The skill facilitates an attack surface for indirect prompt injection as it processes external web content.
Ingestion points: The agent reads untrusted data from the web using open, snapshot, get, and find commands.
Boundary markers: The instructions do not explicitly mandate the use of delimiters or warnings to prevent the agent from following instructions embedded in web content.
Capability inventory: The skill has broad capabilities, including full browser control and the ability to execute arbitrary JavaScript via the eval command.
Sanitization: There is no documentation regarding the sanitization or validation of data retrieved from websites before it is used in downstream logic.

steel-browser