oVirt MCP Server

Manage oVirt/RHV virtualization infrastructure via MCP. Provides 186 tools for VMs, hosts, clusters, networks, storage, templates, snapshots, disks, events,...

This skill is documentation for an oVirt MCP server and is not itself code, but the SKILL.md instructs installing and running external software that requires oVirt credentials. Before installing or running: 1) Verify the upstream package/repository (PyPI package and GitHub repo) and review its source; 2) Treat OVIRT_ENGINE_PASSWORD as a secret — use environment variables or a secrets manager, do not embed passwords in JSON config files; 3) Run the MCP server in an isolated environment (sandbox/container) if you don't fully trust the package; 4) Prefer least-privilege oVirt account for automation (avoid full admin credentials where possible); 5) If you expect automated agent invocation, confirm that exposing a local MCP server and its RPC interface is acceptable in your environment; 6) Ask the skill author or registry maintainer why required env vars and install steps were omitted from the declared metadata — this mismatch is the primary reason for caution.

Type: OpenClaw Skill Name: ovirt-mcp Version: 0.1.0 The ovirt-mcp skill bundle provides 186 tools for comprehensive management of oVirt/RHV virtualization infrastructure, including VM lifecycles, host fencing, and RBAC configuration. While the tools are clearly aligned with the stated purpose and no evidence of intentional malice was found, the bundle contains 'meaningful high-risk behaviors' and 'broad permissions' (e.g., host_install accepting root passwords in host.md, and tools for deleting VMs or modifying permissions in vm.md and rbac.md). According to the provided criteria, these risky capabilities—even when plausibly needed for the stated purpose—warrant a suspicious classification due to the potential for significant impact if the agent is misused.