The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The troubleshooting guide suggests a pattern for downloading and immediately executing a shell script from the vendor's domain to reinstall system components.
Evidence: wget -nv -O- https://lambdalabs.com/install-lambda-stack.sh | sh - in references/troubleshooting.md.
[COMMAND_EXECUTION]: The advanced usage documentation provides Python implementations for executing arbitrary shell commands on remote instances using the SSH protocol.
Evidence: The run_remote_job function in references/advanced-usage.md uses paramiko.SSHClient.exec_command(cmd) to run lists of user-provided or hardcoded commands on instances.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and display output from remote systems and API services, creating a surface for potential instruction injection from untrusted data.
Ingestion points: Remote command output captured via stdout.read() in references/advanced-usage.md, and instance metadata fetched via api.list_instances() in SKILL.md.
Boundary markers: None. The skill does not employ specific delimiters or instructions to ignore potential commands embedded in the remote output.
Capability inventory: The skill possesses capabilities for network operations (Lambda API, SSH), local file writing (checkpoint management), and execution of remote shell commands.
Sanitization: None. Data from remote command execution and API responses is printed directly to the console without validation or escaping.
[EXTERNAL_DOWNLOADS]: The skill relies on and installs several external dependencies for its core functionality.
Evidence: Installation of lambda-cloud-client, paramiko, transformers, and hf_transfer via standard package managers.

lambda-labs-gpu-cloud