lava-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| LAVA_URL | No | LAVA base URL (stdio fallback; HTTP clients send X-Lava-Url) | |
| LAVA_TOKEN | No | API token (stdio fallback; HTTP clients send X-Lava-Token) | |
| LAVA_MCP_HOST | No | HTTP bind host (hosted mode) | |
| LAVA_MCP_PORT | No | HTTP bind port (hosted mode) | |
| LAVA_API_VERSION | No | REST version | v0.3 |
| LAVA_MCP_READ_ONLY | No | Hide write tools (set to 'true' to enable read-only mode) | |
| LAVA_MCP_TRANSPORT | No | Transport type: 'stdio' or 'streamable-http' | stdio |
| LAVA_MCP_GATEWAY_PORT | No | SSH gateway port | 2222 |
| LAVA_MCP_GATEWAY_ENABLED | No | Enable interactive SSH board-session gateway (set to 'true' to enable) | |
| LAVA_MCP_GATEWAY_ADVERTISE_HOST | No | Host containers dial back to |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| whoamiA | Return the LAVA user your token authenticates as. |
| versionA | Return the version of the connected LAVA server. |
| list_devicesA | List devices, optionally filtered by device_type, health or state. Returns {count, results}. health is e.g. Good/Bad/Maintenance/Unknown; state is Idle/Reserved/Running. |
| get_deviceB | Get the full record for one device by hostname. |
| get_device_dictionaryC | Get a device's rendered configuration dictionary (Jinja2/YAML text). |
| get_qdl_infoA | Summarise a device's QDL/flash capability (qdl/fastboot deploy + boot params). Useful before flashing a Qualcomm board: reports whether the device supports qdl, the qdl deploy/boot method parameters, and all available deploy/boot methods, derived from the device's rendered configuration. |
| list_device_typesC | List the device types known to this LAVA instance. |
| list_workersA | List the dispatcher workers and their health/state. |
| list_jobsB | List test jobs, newest first, with optional filters. state is e.g. Submitted/Scheduling/Scheduled/Running/Canceling/Finished; health is Unknown/Complete/Incomplete/Canceled. |
| get_jobA | Get the full record (state, health, device, times) for one job. |
| get_job_definitionA | Get the original submitted YAML job definition for a job. |
| get_job_logsA | Get a job's logs (YAML). Optionally limit to the [start, end) line range. |
| get_job_resultsC | Get a job's test-case results (pass/fail per case). |
| get_queueB | Get the queue of submitted jobs waiting for a device. |
| get_runningB | Get per-device-type running/reserved counts. |
| get_lab_healthB | Get per-device health across the lab. |
| validate_jobB | Validate a YAML job definition without submitting it. |
| submit_jobC | Submit a YAML job definition. Returns the new job id(s). |
| cancel_jobB | Request cancellation of a running or queued job. |
| resubmit_jobC | Resubmit a finished job with the same definition. |
| set_job_priorityB | Set a job's queue priority (0-100, higher runs sooner). |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
| devices_resource | The current device inventory. |
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/mattface/lava-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server