Skip to main content
Glama
codebymov

CarlaMCP

by codebymov
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Local

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault

No arguments

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

CapabilityDetails
tools
{
  "listChanged": false
}
prompts
{
  "listChanged": false
}
resources
{
  "subscribe": false,
  "listChanged": false
}
experimental
{}

Tools

Functions exposed to the LLM to take actions

NameDescription
get_statusA

Check whether the CARLA server is reachable and return version info.

Always call this first before running scenarios.

Returns: dict: carla_available/ready flags plus version info or an error hint describing how to start the simulator.

list_mapsA

List the CARLA maps supported by CarlaMCP.

Returns: dict: A maps mapping of map name to a short description of its characteristics.

run_scenarioA

Generate and execute a CARLA scenario from natural language.

Builds a validated :class:ScenarioManifest from flat parameters and runs it. At least one of the cut_in_*, pedestrian_*, or weather_* groups must be provided. Optionally persists the validated spec for replay.

Args: nl_prompt (str): The natural-language description of the scenario. cut_in_speed_kmh (Optional[float]): See :func:_build_manifest. cut_in_ttc_s (Optional[float]): See :func:_build_manifest. cut_in_side (Optional[str]): See :func:_build_manifest. cut_in_ego_speed_kmh (float): See :func:_build_manifest. cut_in_map (str): See :func:_build_manifest. pedestrian_gap_s (Optional[float]): See :func:_build_manifest. pedestrian_distance_m (Optional[float]): See :func:_build_manifest. pedestrian_occluded (bool): See :func:_build_manifest. weather_condition (Optional[str]): See :func:_build_manifest. weather_intensity (Optional[float]): See :func:_build_manifest. save_as (Optional[str]): If given, save the validated spec under this name for later replay.

Returns: dict: The execution result, or an error/hint payload if validation failed.

Examples: Aggressive cut-in from the right at 120 km/h::

    run_scenario("...", cut_in_speed_kmh=120, cut_in_ttc_s=1.2,
                 cut_in_side="right")

Occluded pedestrian crossing 30 m ahead::

    run_scenario("...", pedestrian_gap_s=1.5, pedestrian_distance_m=30,
                 pedestrian_occluded=True)
set_weatherA

Apply a weather condition to the world without spawning agents.

Args: condition (str): One of clear, rain, fog, night, rain_night. intensity (float): Severity from 0.0 (minimal) to 1.0 (maximum).

Returns: dict: The execution result.

validate_scenarioA

Validate a scenario manifest JSON without executing it.

Args: manifest_json (str): JSON string of a :class:ScenarioManifest.

Returns: dict: {"valid": True, "manifest": ..., "primitives": [...]} on success, or {"valid": False, "error": ...} on failure.

save_scenarioB

Validate and persist a scenario manifest for later replay.

Args: manifest_json (str): JSON string of a :class:ScenarioManifest. name (str): Short name to save under (stored as scenarios/<name>.json).

Returns: dict: {"saved": True, "path": ...} on success, or {"saved": False, "error": ...} if validation failed.

replay_scenarioA

Load a previously saved scenario spec and execute it.

Args: path (str): Saved scenario name or path (e.g. "demo" or "scenarios/demo.json").

Returns: dict: The execution result, or {"error": ...} if the spec could not be loaded.

Prompts

Interactive templates invoked by user choice

NameDescription

No prompts

Resources

Contextual data attached and managed by the client

NameDescription

No resources

Deploy Server
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

TDQS

A4.1/5.0

Scored across 7 tools

Disambiguation5/5

Each tool has a clearly defined purpose with no overlap: server status, map listing, scenario generation from NL, replay, save, weather control, and validation. The pair run_scenario and replay_scenario are distinguished by input source (NL vs. saved spec).

Naming Consistency5/5

All tools follow the verb_noun snake_case convention (get_status, list_maps, run_scenario, etc.), making the API predictable and easy to navigate.

Tool Count5/5

Seven tools cover the essential operations for a CARLA simulation server: status check, map exploration, scenario lifecycle (create, save, replay, validate), and weather control. The count is well-scoped without redundancy.

Completeness4/5

The tool surface covers the core workflow (scenario generation, execution, replay, weather). Minor gaps include no tool to list or delete saved scenarios, but agents can work around by using the file system or naming conventions.

Maintenance

ActivityInactive
ResponsivenessNo issues

Resources