The energy specialist your agent can call

Help households plan solar, weatherization, EV charging, batteries and heat pumps. Your agent gets planning estimates, incentive sources, quote terms and practical questions to bring back to the person making the decision.

12 tools need no key to start. Informational tools share 20 anonymous calls per caller every 24 hours; a free key includes 100 calls per rolling 30 days and three eligible commercial trials. Human review and installer introductions are optional follow-through.

Choose the job in front of your user

Start with one of these free calls. Examples use fictional inputs; replace them with your user's authorized facts.

estimate_production({
  "zipCode": "59715",
  "country": "US",
  "systemKw": 8
})

list_guides({
  "region": "Montana",
  "topic": "weatherization",
  "limit": 3
})

get_node_score({
  "zipCode": "59715",
  "country": "US",
  "serviceType": "ev_charger",
  "evType": "own_ev",
  "electricalPanelSize": "unknown"
})

get_node_score({
  "zipCode": "59715",
  "country": "US",
  "serviceType": "battery",
  "backupNeed": "essentials",
  "outageConcern": "occasional"
})

list_guides({
  "region": "Montana",
  "topic": "heat_pump",
  "limit": 3
})

US guides and installer routing. Non-US incentive guidance and production estimates use disclosed national assumptions. Check source dates and program eligibility before relying on incentive guidance. Agents can read the same capability map directly, or open energyai://capabilities through MCP resources.

Have a quote? Return a useful first review

{
  "jsonrpc": "2.0", "id": 1, "method": "tools/call",
  "params": {
    "name": "review_installer_quote",
    "_meta": { "toolCallId": "your-unique-case-id", "maxPriceCents": 10 },
    "arguments": {
  "postalCode": "59715",
  "serviceType": "solar",
  "monthlyBillRange": "150_250",
  "quoteText": "$31,800 cash for 8.0 kW, 20 panels, string inverter, 25-year panel warranty"
    }
  }
}

This automated first pass uses redacted text up to 6,000 characters. It does not establish price fairness, incentive eligibility or equipment suitability. The $49 human-reviewed brief is a separate service. Attach your key as a Bearer header. Check live pricing and use an operator-approved price ceiling; reuse the request ID only for an identical retry. After the operator evaluates a result, use record_quote_review_outcome to record their actual feedback. A successful call alone does not demonstrate customer value.

Discovery tools — no key needed to start

Anonymous informational allowance: 20 calls/caller/24h; free-key allowance: 100 calls/30d. Sustained informational use then requires an active Builder subscription; prepaid credit is for priced commercial tools. get_power_service_quote, get_power_passport_link, get_quote_link, route_lead, find_local_installers, create_builder_key, and get_builder_upgrade_link stay always free without a key. Hourly safety limits: 120 calls/caller (route_lead 10/hour). Machine-readable catalog: /api/v1/agent

Commercial tools — trial, subscribe, or stay prepaid

New keys include three trial calls for tools priced at $0.25 or less. Builder is $19/month with $20 in monthly tool credit; prepaid top-ups remain available with a $5 minimum. Failed billed calls and locked roadmap previews do not consume credit. Live machine-readable pricing: /api/v1/mcp/pricing

Connect via MCP

{ "mcpServers": {
  "energyai": { "type": "http", "url": "https://api.energyaisolution.com/mcp" }
}}

claude mcp add --transport http energyai https://api.energyaisolution.com/mcp

Or plain REST

curl -X POST https://api.energyaisolution.com/api/v1/agent/check_incentives \
  -H 'content-type: application/json' \
  -d '{"args":{"zipCode":"59715"}}'

OpenAI function-calling / LangChain

GET https://api.energyaisolution.com/api/v1/agent/openai-tools.json
→ { "tools": [ { "type": "function", "function": { "name": "check_incentives", ... } }, ... ] }

import requests

def call_energyai(name: str, args: dict):
    r = requests.post(f"https://api.energyaisolution.com/api/v1/agent/{name}",
                      json={"args": args}, timeout=30)
    return r.json()  # { ok, data } | { ok: False, error }

call_energyai("estimate_production", {"zipCode": "85004", "systemKw": 8})

Routing a homeowner (consent required)

Power Passport — specialist site decision service

get_power_passport_link({
  "siteLocation": "Clark County, Nevada",
  "targetMw": 12,
  "workloadType": "inference",
  "projectStage": "utility_conversations",
  "decisionNeed": "Decide whether to stage the first 4 MW here or continue the site search."
})

Power Screen — the agent-to-agent purchase

const quote = get_power_service_quote({
  "siteLocation": "Clark County, Nevada",
  "targetMw": 12,
  "workloadType": "inference",
  "flexibilityHours": 24,
  "stagedMw": 4,
  "coolingMode": "dry",
  "powerEvidence": "utility_contact"
})

run_power_screen({ "quoteToken": quote.quote.quoteToken })

Why EnergyAI

EnergyAI keeps supplied facts, modeled estimates and missing evidence visible. Quote terms are checked against the submitted text; sources and assumptions travel with the result. Your agent can deliver that result in the conversation, and the household can choose to keep a private decision record. A tool call alone does not establish savings or a successful installation.