--- name: unloopa-api description: Make your agent sell websites to local businesses on autopilot. Finds leads from Google Maps, builds a custom AI website for each one, sends outreach emails, and can even call them. Use when the user wants to find leads, generate websites, send emails, or make voice calls. metadata: author: unloopa version: "1.3" openclaw: primaryEnv: UNLOOPA_API_KEY requires: env: - UNLOOPA_API_KEY ---

Unloopa API

You control the Unloopa platform through its REST API. All requests go to https://dashboard.unloopa.com/api/v1/ with Bearer token authentication.

Authentication

Every request needs the header:

Authorization: Bearer $UNLOOPA_API_KEY

The API key is set in the UNLOOPA_API_KEY environment variable. Keys start with unlpa_live_.

If the user hasn't configured their key yet, tell them:

1. Go to https://dashboard.unloopa.com/settings and click the "API" tab
2. Click "Create API Key" and copy the key (it's only shown once)
3. Configure it in your OpenClaw settings, or set the environment variable: export UNLOOPA_API_KEY=unlpa_live_...

If you get a 401 unauthorized error, the key is missing or invalid — ask the user to check their key.

First Call: Always Start With GET /quota

Before doing anything else, call GET /quota to discover:

• What plan the user is on (starter or pro)
• Whether voice and video features are enabled
• How many leads and credits remain
• Purchase links if they need to upgrade or buy credits

This single call tells you everything about what the user can and can't do. Adapt your behavior based on the response:

| quota field | What it means | |---|---| | voice_enabled: false | Don't offer voice calling — they need Pro plan. Share purchase_links.upgrade | | video_enabled: false | Don't offer video generation — they need Pro plan | | voice_credits: 0 | Can't make calls — share purchase_links.voice_credits | | websites.remaining: 0 | Can't generate leads — quota resets at resets_at |

Error Format

All errors return:

{ "error": { "code": "error_code", "message": "Human-readable message", "details": {} } }

Error codes: unauthorized (401), invalid_input (400), not_found (404), plan_required (403), insufficient_credits (402), quota_exceeded (429), rate_limited (429), setup_required (400), limit_reached (400), invalid_state (400), internal_error (500).

When you get plan_required (403), share the upgrade link from quota. When you get insufficient_credits (402), share the credit purchase links. When rate limited, check the Retry-After header (seconds).

Plans

• Starter ($47/mo): 1,000 leads/month, email outreach, templates
• Pro ($97/mo): 5,000 leads/month, 200 videos/month, voice calling, AI agents, phone numbers

---

Workflows

1. Full Lead Pipeline (any plan)

The /command endpoint runs the entire pipeline automatically: scrape leads → generate websites → enrich emails → send outreach. Just describe what you want.

1. GET  /quota                    → check websites.remaining > 0
2. POST /command                  → submit natural language command (full pipeline runs automatically)
3. GET  /jobs/{job_id}            → poll every 5-10s until status=completed
4. GET  /leads?job_id={job_id}    → view generated leads with websites, emails, etc.

2. Email Outreach (any plan)

1. GET  /outreach/status          → verify configured=true, remaining_today > 0
2. GET  /leads?has_email=true     → find leads with emails
3. GET  /outreach/templates       → pick a template
4. POST /outreach/send            → queue emails

If configured=false, tell the user to connect an email account at the setup_url in the response.

3. Voice Calling (Pro plan only)

Skip this workflow entirely if voice_enabled=false in /quota. Tell the user they need Pro and share the upgrade link.

Prerequisites: voice_enabled=true + voice_credits > 0 + at least 1 phone number + at least 1 voice agent.

1. GET  /quota                    → voice_enabled? voice_credits > 0?
   If voice_credits=0 → share purchase_links.voice_credits
2. GET  /phone-numbers            → count > 0? (max 3)
   If empty → POST /phone-numbers/search + POST /phone-numbers/buy
3. GET  /voice/agents             → count > 0? (max 3)
   If empty → POST /voice/agents (create one)
4. POST /voice/call               → single call, OR:
5. POST /voice/campaigns          → bulk campaign (starts as draft)
6. PATCH /voice/campaigns/{id}    → action=activate, then action=trigger

4. Full Funnel

The /command endpoint now handles steps 1-3 automatically. Voice calling is the only manual step.

1. GET  /quota                    → know the plan, adapt accordingly
2. POST /command → poll /jobs/{id} → GET /leads  (scrape + websites + emails + outreach all automatic)
3. Voice (Pro only): /phone-numbers → /voice/agents → /voice/campaigns

---

Endpoints Reference

POST /command

Submit a natural language lead generation command. The API automatically runs the full pipeline: scrape → generate websites → enrich emails/socials → send outreach. No need to mention each step in the command.

Body:

{
  "command": "Find 50 plumbers in Miami",
  "max_results": 50,
  "with_video": false,
  "with_vsl": false
}

• command (required, string, max 1000 chars) — just describe the niche and location. Any number mentioned in the command is ignored — use max_results to control lead count.
• max_results (optional, 1-100, default: 100, or 10 when with_video/with_vsl is true)
• with_video (optional, bool, Pro plan only)
• with_vsl (optional, bool, Pro plan only)

Default behavior: The API always overrides what's in the command text. It scrapes up to max_results leads (default 100), generates a website for each, finds email addresses, enriches social profiles, and sends outreach emails — all automatically. Numbers in the command like "Find 15 plumbers" are ignored; use max_results instead.

Response: { job_id, status: "processing", defaults: { max_results, generate_websites, enrich_emails, send_outreach, with_video, with_vsl }, quota: { used, limit, remaining } }

---

GET /jobs

List submitted commands.

Query: ?limit=20&offset=0 (limit max 100)

Response: { jobs: [{ job_id, command, intent, status, error, created_at, updated_at }], total, limit, offset }

---

GET /jobs/{id}

Poll a job for progress and results.

Response:

{
  "job_id": "uuid",
  "status": "processing|completed|failed",
  "progress": 75,
  "current_step": "Generating websites...",
  "steps": [{ "name": "scraping", "status": "completed", "message": "Found 50 leads", "count": 50 }],
  "result": {
    "websites": [{ "id": "uuid", "url": "https://...", "business_name": "...", "city": "...", "industry": "..." }],
    "leads_scraped": 50,
    "emails_sent": 0
  },
  "error": null
}

Poll every 5-10 seconds. Jobs take 30s to 5 minutes depending on count and video.

---

GET /leads

List and filter leads.

Query params (all optional):

• limit (1-100, default 50), offset (default 0)
• city — partial match (e.g. "miami")
• industry — partial match (e.g. "plumber")
• has_phone=true — only leads with phone numbers
• has_email=true — only leads with email addresses
• min_rating — minimum Google rating (e.g. 4.0)
• min_reviews — minimum review count
• job_id — leads from a specific command
• search — free text search across name, city, industry
• created_after — ISO date (e.g. "2025-01-15")
• created_before — ISO date
• has_website=true — only leads with generated website URLs
• has_video=true — only leads with video
• video_status — pending|generating|completed|failed

...