Agentsumo Mcp
by mw-jeongcommunityv0.1.2
1
AgentSUMO is an MCP server and framework for generating, running, and analyzing SUMO traffic simulation scenarios from natural-language prompts. It translates policy questions into executable simulation plans and supports a web dashboard for visualization and cross-scenario analysis.
From the registry: MCP server for LLM-driven SUMO traffic simulation, scenario design, and result analysis.
Install & configure
Prerequisites
- requiredInstall Python 3.10+Python 3.10 or later is required to run the AgentSUMO MCP server.https://www.python.org/downloads/
- requiredInstall SUMO 1.24+Install SUMO locally. The MCP server requires a local SUMO installation and the SUMO_HOME environment variable must point to that installation directory containing bin/sumo or bin/sumo.exe.https://eclipse.dev/sumo/
- requiredSet SUMO_HOMESet the SUMO_HOME environment variable to the absolute path of your local SUMO installation.
- optionalInstall uvx or pipUse uvx to run the published MCP server without installation, or install the package with pip. The README presents uvx as a supported standalone path and pip install as an alternative.https://docs.astral.sh/uv/guides/tools/
- requiredCreate an Anthropic API keyThe server/framework documentation states ANTHROPIC_API_KEY is required. This Claude API key is used by AgentSUMO.https://console.anthropic.com/settings/keys
Install in your client
Paste into the AI you're chatting with — it will figure out the rest
Please install the `agentsumo-mcp` MCP server into my current AI client (that's you).
Required prerequisites (do these first if not already done):
- **Install Python 3.10+** — Python 3.10 or later is required to run the AgentSUMO MCP server. (https://www.python.org/downloads/)
- **Install SUMO 1.24+** — Install SUMO locally. The MCP server requires a local SUMO installation and the SUMO_HOME environment variable must point to that installation directory containing bin/sumo or bin/sumo.exe. (https://eclipse.dev/sumo/)
- **Set SUMO_HOME** — Set the SUMO_HOME environment variable to the absolute path of your local SUMO installation.
- **Create an Anthropic API key** — The server/framework documentation states ANTHROPIC_API_KEY is required. This Claude API key is used by AgentSUMO. (https://console.anthropic.com/settings/keys)
Optional prerequisites:
- Install uvx or pip — Use uvx to run the published MCP server without installation, or install the package with pip. The README presents uvx as a supported standalone path and pip install as an alternative. (https://docs.astral.sh/uv/guides/tools/)
Canonical MCP server config (stdio transport):
- command: `uvx`
- args: ["agentsumo-mcp"]
- required environment variables:
- `ANTHROPIC_API_KEY`: Anthropic Claude API key used by AgentSUMO. (example: `<your-anthropic-api-key>`)
- `SUMO_HOME`: Absolute path to the local SUMO installation directory containing bin/sumo or bin/sumo.exe. (example: `<path-to-your-sumo-installation>`)
- optional environment variables:
- `AGENTSUMO_MCP_OUTPUT_BASE`: Optional base directory where the MCP server writes simulation outputs such as networks, trips, and results. Defaults to the current working directory. (example: `<path-to-your-output-directory>`)
- `MAPBOX_TOKEN`: Mapbox access token used by the web UI renderer. Not required for the standalone MCP server unless your workflow depends on web UI features. (example: `<your-mapbox-token>`)
Note: Prefer the standalone published server path: run with uvx agentsumo-mcp, or alternatively install with pip install agentsumo-mcp and invoke the installed command. The README does not provide a client config snippet with additional args, so the canonical stdio shape is command `uvx` with args `["agentsumo-mcp"]`.
Add this MCP server to my current client's config in the correct format for you. If you need secrets or credentials I haven't provided, ASK me — do not invent values or leave raw placeholders. After adding it, tell me how to verify the server is connected.Note: Prefer the standalone published server path: run with uvx agentsumo-mcp, or alternatively install with pip install agentsumo-mcp and invoke the installed command. The README does not provide a client config snippet with additional args, so the canonical stdio shape is command `uvx` with args `["agentsumo-mcp"]`.
Tools (21)
osm_extract
Extract OpenStreetMap data for SUMO network creation
net_convert
Convert extracted OSM data into a SUMO network
trip_generate
Generate trips for a simulation scenario
route_generate
Generate routes from trips or network data
sumo_runner
Run a SUMO simulation
edge_edit_tool
Modify road edges for infrastructure changes
reduce_lanes_tool
Reduce the number of lanes on edges
vehicle_generation_tool
Generate vehicles for demand or policy experiments
flow_generation_tool
Generate traffic flows for demand scenarios
tls_offset_tool
Adjust traffic light offsets
tls_adaptation_tool
Adapt traffic signal settings
xml_to_sqlite_tool
Convert SUMO XML outputs into a SQLite database
simulation_report_tool
Generate an HTML report from simulation results
visualize_net_tool
Environment variables
ANTHROPIC_API_KEYrequiredClaude API key used by the Planner AgentMAPBOX_TOKENrequiredMapbox access token used to render the web map basemapSUMO_HOMErequiredAbsolute path to the local SUMO installation, used to locate SUMO binariesPrerequisites
- • Python 3.10+
- • Python 3.10
- • Python 3.12
- • python 3.12
Packages
agentsumo-mcp
pypiv0.1.2transport: stdio
Details
- Language
- Python
- License
- MIT
- Released
- Jun 5, 2026
- Last commit
- Jun 5, 2026
- Registry name
- io.github.mw-jeong/agentsumo-mcp
Categories
Related Servers
Claude Flow
AI orchestration with hive-mind swarms, neural networks, and 87 MCP tools for enterprise dev.
64.6k
Codebase Memory Mcp
Persistent codebase knowledge graph. Survives session restarts and context compaction.
31.9k
Serena
A powerful toolkit for coding, providing semantic retrieval and editing capabilities.
26.5k