This skill informs the agent how to use Viam CLI

Provides comprehensive guidance and commands to install, authenticate, and manage Viam CLI for robotics fleets, modules, machine data, datasets, and ML workf...

This skill is an instruction-only guide for the Viam CLI and appears internally consistent. Before installing or allowing any operations: (1) only permit the Homebrew install if you trust the source and you want the Viam CLI on your machine; (2) when the agent requests to perform destructive actions (delete, create API keys, open remote shells, copy files), confirm each action explicitly and review the exact command it will run; (3) do not paste secrets into chat — use viam login in your terminal or your platform's environment injection as recommended; (4) note the SKILL.md included a requirement block (anyBins: ["viam","brew"]) even though the registry metadata listed none — this minor mismatch is likely benign but worth noting; (5) if you need higher assurance, ask the skill author/source or request the full, untruncated SKILL.md (the provided content was truncated) before proceeding.

Type: OpenClaw Skill Name: viam-cli Version: 1.0.0 The skill is classified as suspicious due to the explicit instructions for the AI agent to utilize high-risk Viam CLI commands such as `viam machines part shell` (remote shell access) and `viam machines part cp` (remote file transfer). While these are legitimate functions of the Viam CLI and the skill includes guardrails like requiring user confirmation for destructive operations and installation, their exposure to an AI agent creates a significant vulnerability to prompt injection, potentially leading to remote code execution, unauthorized data exfiltration, or code deployment if the agent is compromised. These capabilities, detailed in `SKILL.md`, represent a critical security risk, even without explicit malicious intent within the skill's design.