netlab-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| NETLAB_MCP_STORE | No | Store directory for the matrix db + artifacts (default: ./store) | |
| NETLAB_MCP_WORKDIR | No | Base directory for per-request temp workdirs (default: ./.work) | |
| NETLAB_MCP_NETLAB_BIN | No | Path to the netlab executable (default: same venv, then PATH) | |
| NETLAB_MCP_ACCEPT_CEOS_EULA | No | Set truthy to allow the EULA-gated ceos image |
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": true
} |
| logging | {} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| extensions | {
"io.modelcontextprotocol/ui": {}
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| generate_topologyA | Turn an intent + target platforms into a netlab topology YAML (parse-validated). intent: free text, e.g. "ebgp peering" or "ospf two routers". The module is inferred.
platforms: NOS list, dut first (MVP free set: srlinux, frr, cumulus, vyos, linux).
Feed the returned |
| render_configA | Render real per-device config from a netlab topology — offline, no containers. Returns {per_node: {node: {module: config_text}}, clab_yaml, disclaimer}. This is the netlab data-model transform + Jinja2 render; the config matches what would deploy. |
| query_compatibilityC | What netlab declares a platform supports, overlaid with what was observed in the lab.
|
| get_known_goodB | Return a previously lab-passed topology + rendered config for module+platform, if any. |
| list_examplesA | Index netlab's integration test topologies (real, maintained multi-platform scenarios). With no module: list available modules + counts. With a module: list its scenarios. |
| report_failureC | Record a negative result (a combo that did not work) into the compatibility matrix. |
| host_checkA | Diagnose this host's lab readiness in one call — run this first when anything fails. Reports docker/containerlab availability + versions, the netlab version, which platforms are allowed, which devices have locally loaded images (deployable without a pull), and which devices can anchor validate tests per module. |
| validate_in_labA | Deploy a topology to containerlab, run Requires docker + containerlab on a Linux host. Returns the verdict (pass/fail/warning), rendered config, raw validate output, and persists a version-scoped matrix row. The lab is always torn down afterward. On a host without containerlab this returns verdict "unavailable" rather than failing. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 8 tools
Each tool has a clearly distinct purpose: topology generation, known-good retrieval, host check, example listing, compatibility query, config rendering, failure reporting, and lab validation. No overlap in functionality.
All tool names follow a consistent verb_noun pattern in snake_case (e.g., generate_topology, validate_in_lab, report_failure). Naming is predictable and readable.
With 8 tools, the set is well-scoped for a netlab MCP server. Each tool is justified and covers a specific step in the workflow without being excessive or insufficient.
The tool surface covers the full lifecycle: topology generation, config rendering, validation, host readiness, compatibility queries, failure reporting, and example browsing. No obvious missing operations for the intended domain.