Early access

When a shopper asks your AI what they can save, it should know.

EV Life’s MCP server gives any AI agent the same savings math that runs on Tier 1 OEM retail sites. Incentives, fuel, charging, and range, callable as tools, resolved in real time for the person asking.

Request access

Why this matters now

Agents are answering car questions. This is the one they get wrong.

A general model can tell a shopper what an EV costs, what it tows, and how far it goes. Ask it what they’ll actually pay after incentives and it guesses, because the answer doesn’t exist until it’s computed against that person’s income, ZIP code, household, utility, and vehicle.

28% of the 500+ US incentive programs we track changed materially in the last 90 days. A model trained six months ago is confidently wrong, and a wrong incentive number isn’t a bad answer, it’s a number a shopper may act on.

So we made the engine callable. Same data, same eligibility logic, same daily verification that four Tier 1 automakers run on their retail sites, available as tools an agent can invoke mid-conversation.

What your AI can answer

Give your AI accurate answers on incentives, savings, range, and charging.

get_incentives

Every federal, state, utility, and municipal program the shopper qualifies for, with amounts, deadlines, and links to apply. Includes charger rebates, HOV decals, and toll discounts.

zip · income · household_size · filing_status · vehicle · utility

get_savings

Fuel, maintenance, and total cost of ownership over a chosen term, using local electricity rates rather than a national average.

zip · vehicle · annual_miles · term_years

get_charging

Charge times by level, cost per full charge at home and on DC fast, and what the vehicle supports.

vehicle · zip · charger_level

get_range

Real-world range with seasonal adjustment, plus trip planning with charging stops, arrival state of charge, and total charge cost.

vehicle · origin · destination · season

Tool names shown are illustrative. Final naming is confirmed during integration.

What it looks like in a conversation

The agent stops guessing and starts calling.

YOU

I make about $95k, family of four, ZIP 10001. What would an IONIQ 5 actually cost me?

AI
→ evlife.get_incentives({zip: "10001", income: 95000, household_size: 4, vehicle: "hyundai-ioniq-5" })

You qualify for $4,200 in incentives on the IONIQ 5, and there are two you probably haven’t seen:

NY Drive Clean Rebate
$2,000
Con Edison charger rebate
$1,500
NYC parking tax exemption
$700
Fuel savings
$150/mo
Est. 5-year savings
$13,200

Where it runs

Anywhere a shopper is already asking.

OEM shopping assistants

The chat on your vehicle pages answers the savings question with the same numbers as your calculator, instead of a disclaimer.

Dealer voice agents

A shopper calls after hours and asks about rebates. The agent answers with real figures for their ZIP rather than taking a message.

In-car and owner apps

Charger rebates, time-of-use rates, and trip costs, surfaced to an owner who never visited a website to find them.

Prefer to build your own interface? See the EV incentives API.

Connecting

Three steps, no data migration.

  1. 1

    Request access. We issue a key scoped to your brands and vehicles.

  2. 2

    Point your agent at the server. Add mcp.evlife.co to your client config. Standard MCP, so anything that speaks it works.

  3. 3

    Ask it a question. The tools appear automatically. Nothing to train, nothing to sync, and the data stays current because we maintain it.

In pilot with integration partners.

We’re onboarding a small number of OEM and dealer platform teams.

Request access