lava
Automotive Grade Linux publishes a first-party MCP server - lava-mcp-toolkit - in its own GitHub organization (Apache-2.0, first release 0.1.0 on 2026-08-27). It exposes the AGL LAVA test infrastructure at https://lava.automotivelinux.org to agents as MCP tools, and is implemented four times over (Go, Python, Rust, C++), each in a `stdio` single-identity variant and a `server` multi-tenant HTTP variant. AGL does NOT operate a hosted/remote MCP endpoint - every variant is run by the consumer on their own machine or their own server, so an agent cannot reach AGL's LAVA through MCP without someone installing and running this first.
One-click install for Cursor, VS Code, Claude, and 20+ other MCP clients, powered by API Commons MCP Install — visit install.apicommons.org for more information.
Documentation
Documentation link · transport
Tools
lava_list_jobs— List LAVA test jobs. Optional filters: state, health, device_type, submitter, ordering, limit, offset.lava_get_job— Get details of a single LAVA job by id.lava_job_logs— Fetch log lines for a job, optionally starting at a given line offset.lava_submit_job— Submit a new LAVA job. definition must be a complete LAVA job YAML string.lava_cancel_job— Cancel a running or queued LAVA job.lava_list_devices— List LAVA devices. Optional filters: device_type, health, state.lava_get_device— Get details of a single device by hostname.lava_raw_request— Escape hatch: raw REST call under /api/v0.2/. method GET|POST|PUT, path relative to /api/v0.2/.lava_whoami— Report which LAVA identity the calling MCP token is mapped to, without revealing the LAVA token.
About MCP
The Model Context Protocol (MCP) is an open protocol Anthropic introduced for connecting LLM-based agents to external tools and data sources. Providers publish MCP servers that expose their API surface as structured, discoverable tools — an MCP-compatible client (Claude Desktop, Cursor, Cline, Continue, etc.) can connect to the server and call its tools without any per-provider integration code.
Browse every MCP server on the APIs.io network or compare with the broader Agent Skill surfaces of the same providers.
Work with this as data
Every MCP server here is available over the APIs.io API and to AI agents over MCP.