Skip to content
VoLCA Docs

Quick Start

VoLCA is one LCA engine with several entry points. This page gives a first usable step for each one. For the decision map, see Ways to use VoLCA.

Hosted product

Section titled “Hosted product”

Use this when you want the fastest start and do not want to install the engine.

  1. Create an account.
  2. Start your managed VoLCA instance from the account dashboard.
  3. Use the browser UI to load or inspect data.

Best first outcome: you can explore VoLCA from the browser without managing a local server.

Desktop app

Section titled “Desktop app”

Use this when you want a local graphical workflow.

  1. Go to Desktop.
  2. Create an account if needed.
  3. Download the app for Windows, macOS, or Linux from your account.
  4. Open the app and point it at your local data.

Best first outcome: you can inspect LCA data locally without starting from terminal commands.

Self-hosted binary

Section titled “Self-hosted binary”

Use this when you want to run the engine yourself.

Download the latest binary from GitHub releases, or build from source:

Terminal window
git clone https://github.com/ccomb/volca
cd volca
cabal build exe:volca

Create a minimal volca.toml in your project directory:

[server]
port = 8080
# password = "secret"   # optional


[[databases]]
name = "ecoinvent"
format = "ecospold2"
path = "/data/ecoinvent-3.10-cutoff"
load = true


[[methods]]
name = "EF3.1"
path = "/data/EF3.1"
load = true

Start the engine:

Terminal window
volca --config volca.toml server --port 8080

Best first outcome: a local VoLCA server is running at http://localhost:8080.

CLI

Section titled “CLI”

Use this when you want scriptable terminal commands.

Search activities directly from the command line:

Terminal window
volca --config volca.toml --db ecoinvent activities --name "electricity"

Then inspect inventory or impacts:

Terminal window
volca --config volca.toml --db ecoinvent inventory <PROCESS_ID>
volca --config volca.toml --db ecoinvent impacts <PROCESS_ID> --method <METHOD_UUID>

Best first outcome: you can get machine-readable or human-readable answers from shell commands. See CLI overview.

REPL

Section titled “REPL”

Use this when you want interactive exploration before writing scripts.

Terminal window
volca --config volca.toml repl

Example session:

volca[ecoinvent]> activities --name electricity
volca[ecoinvent]> inventory <UUID>
volca[ecoinvent]> :format json
volca[ecoinvent]> impacts <UUID> --method <METHOD_UUID>
volca[ecoinvent]> :quit

Best first outcome: you can explore a database with session state and tab completion. See REPL reference.

HTTP API

Section titled “HTTP API”

Use this when another application, notebook, dashboard, or pipeline needs to call VoLCA.

Start a server, then call the API under /api/v1/:

Terminal window
curl "http://localhost:8080/api/v1/db/ecoinvent/activities?name=electricity&limit=20"

If your config sets a password, authenticate with a bearer token:

Terminal window
curl -H "Authorization: Bearer yourpassword" \
  "http://localhost:8080/api/v1/db/ecoinvent/activities?name=electricity&limit=20"

Best first outcome: your software can query loaded LCA data over HTTP. See HTTP API reference.

Python client: pyvolca

Section titled “Python client: pyvolca”

Use this when your workflow is already in Python.

Install the client:

Terminal window
pip install pyvolca

Connect to an existing hosted or self-hosted VoLCA server:

from volca import Client


client = Client(base_url="http://localhost:8080", db="ecoinvent", password="yourpassword")
activities = client.search_activities(name="electricity", limit=20)
print(activities[0])

Use download() and Server only if you deliberately want Python to download and launch a local VoLCA engine process for you.

Best first outcome: Python can query VoLCA without manually parsing CLI output. See pyvolca.

MCP

Section titled “MCP”

Use this when you want Claude, Cursor, ChatGPT, or another MCP client to work with VoLCA data through tools.

Start a VoLCA server, then connect your MCP client to:

http://localhost:8080/mcp

Use bearer-token authentication if your server has a password:

Authorization: Bearer yourpassword

Best first outcome: an AI agent can search activities, inspect inventories, and compute impacts through VoLCA tools. See MCP overview.

Next steps

Section titled “Next steps”