RestoAPI · for developers & AI agents
The API for venues about to open
RestoAPI is the self-serve, technical surface for RestoSignals: a hosted MCP server, a REST endpoint, and an installable skill — one scored /v1/openings call, 1 credit = 1 returned lead, empty results free. Start free, pay as you go. No subscription needed.
Developer pricing · pay as you go
Free key
Top-up
Top-up
Metered per returned lead — filters that return nothing cost nothing. Packs never expire. Need the dashboard, contacts and territory tiers instead? See RestoSignals plans (Pro bundles full API access).
Connect the MCP server
The server speaks Model Context Protocol (JSON-RPC 2.0) at https://api.restosignals.com/mcp. Clients that support OAuth (Claude.ai and ChatGPT connectors, Claude Code, Cursor) need only the URL and a one-time sign-in. Or drop this block into any MCP client and swap in your key:
{
"mcpServers": {
"restosignals": {
"url": "https://api.restosignals.com/mcp",
"headers": { "Authorization": "Bearer pk_YOUR_KEY" }
}
}
} Claude.ai / ChatGPT connectors
Add a custom connector with the URL https://api.restosignals.com/mcp. OAuth runs automatically: sign in once (or create a free account) and approve.
Claude Desktop / Code
Add to your MCP config (claude_desktop_config.json, or via the Claude Code MCP settings). URL-only uses OAuth; add the header to use a key.
Cursor
Settings → MCP → Add server, same JSON block.
n8n / LangChain / custom
Point any MCP-over-HTTP client at the URL with the Bearer header, or call the JSON-RPC endpoint directly at https://api.restosignals.com/mcp.
The tools
Four tools, same wallet as REST. Always call coverage first so you query a state that actually has data.
Live states + how many scored venues (and hot leads, score ≥ 80) each holds. Call this FIRST so you never query a state with no data.
{ /* no arguments */ }{
"states": [
{ "state": "NY", "venues": 812, "hot_leads": 41 },
{ "state": "IL", "venues": 190, "hot_leads": 6 },
{ "state": "TX", "venues": 5300, "hot_leads": 0 }
]
}Scored, fused venue leads for a state, ranked by opening_score. signal_types are liquor / food_permit / business.
{
"state": "NY", // required, 2-letter
"min_score": 40, // default 30
"limit": 10 // default 10, max 100
}{
"count": 1,
"openings": [
{ "id": 4412, "business_name": "DINER 24 BROADWAY NYC LLC.",
"city": "New York", "state": "NY", "opening_score": 100,
"status": "opening", "signal_types": ["business","food_permit","liquor"] }
]
}The raw underlying signals (liquor / permit / filing rows) that were fused into one venue — show your work.
{ "venue_id": 4412 }{
"venue": { "id": 4412, "business_name": "DINER 24 …", "opening_score": 100, "…": "…" },
"signals": [
{ "source": "ny_sla", "signal_type": "liquor", "status": "pending", "…": "…" },
{ "source": "nyc_permits", "signal_type": "food_permit", "…": "…" }
]
}Register a webhook that fires when a new matching venue surfaces in an area.
{
"url": "https://you.example.com/hook",
"criteria": { "state": "NY", "city": "Brooklyn",
"signal_type": "liquor", "min_opening_score": 60 }
}{ "webhook_id": 7, "url": "…", "criteria": {…},
"note": "New matching venues will be POSTed here on the daily run." }Or install the skill
A packaged Agent Skill that teaches Claude exactly when and how to use RestoSignals — the coverage-first rule, the real signal_types, and honest coverage so the agent never promises a state we don't have.
↓ Download SKILL.md- Claude Code — drop SKILL.md into .claude/skills/restosignals/ in your project (or ~/.claude/skills/ for all projects).
- claude.ai — add it as a custom skill in your workspace's skill settings.
- Any agent — paste its contents into your system prompt / tool instructions.
Recipes
Prefer straight REST?
# 1) mint a trial key
curl -s -X POST https://api.restosignals.com/v1/register # -> pk_...
# 2) pull scored leads (see /docs for all params)
curl -s "https://api.restosignals.com/v1/openings?state=NY&min_opening_score=40&limit=5" \
-H "X-API-Key: pk_..." Full endpoint reference, params and response fields are in the API docs. Machine-readable summary for crawlers: /llms.txt.