The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill installs the truefoundry Python package (pinned to version 0.5.0) using pip or uv. This is a standard installation procedure for the vendor's official command-line interface.
[COMMAND_EXECUTION]: The skill executes shell commands, specifically tfy login, which incorporate a user-supplied tenant URL as a command-line argument. While the instruction template wraps the input in double quotes, this remains a potential surface for command injection if the agent does not strictly validate the input.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input in the form of a tenant URL provided by the user during the registration flow.
Ingestion points: User-supplied tenant URL in Step 3 and Step 4.
Boundary markers: The instructions lack explicit boundary markers or "ignore embedded instructions" warnings for the user-provided URL.
Capability inventory: The skill allows shell command execution (tfy, bash), package installation (pip, uv), and Python execution via the allowed-tools configuration.
Sanitization: There is no defined sanitization or validation logic to check the format or safety of the tenant URL before interpolation into shell commands.
[DYNAMIC_EXECUTION]: An inline Python script is utilized to inspect the contents of ~/.truefoundry/credentials.json to verify the user's login status. The script is hardcoded within the skill and does not execute untrusted logic, serving as a functional check for local configuration.

truefoundry-onboard