> ## Documentation Index
> Fetch the complete documentation index at: https://docs.mavera.io/llms.txt
> Use this file to discover all available pages before exploring further.

# Smart View Content Generation

> Pull leads from Close CRM Smart Views and auto-generate personalized outreach content using Mavera's Generations API

## Scenario

Close's Smart Views are saved lead filters ("Enterprise FinTech", "Churned Q4"). You want to **pull leads matching a Smart View and auto-generate personalized outreach content** using Mavera's Generations API, tailored to that segment.

## Architecture

```mermaid theme={"dark"}
flowchart LR
    A[Close Smart Views] --> B[Fetch matching leads] --> C["POST /api/v1/generations"] --> D[Personalized outreach content]
```

## Code

<CodeGroup>
  ```python Python theme={"dark"}
  import os, requests
  from requests.auth import HTTPBasicAuth

  CLOSE_KEY = os.environ["CLOSE_API_KEY"]
  CLOSE_AUTH = HTTPBasicAuth(CLOSE_KEY, "")
  MAVERA_KEY = os.environ["MAVERA_API_KEY"]

  def close_get(path, params=None):
      r = requests.get(f"https://api.close.com/api/v1{path}", auth=CLOSE_AUTH, params=params or {})
      r.raise_for_status()
      return r.json()

  # 1. List Smart Views
  views = close_get("/saved_search/", {"_type": "lead"}).get("data", [])
  print("Smart Views:")
  for v in views:
      print(f"  - {v['name']} (id: {v['id']})")

  target = views[0] if views else None
  if not target:
      print("No Smart Views found.")
      exit()

  # 2. Fetch matching leads
  leads = close_get("/lead/", {"query": target.get("query", ""), "_limit": 50,
      "_fields": "id,display_name,contacts,custom,description"}).get("data", [])
  print(f"\nUsing: {target['name']} ({len(leads)} leads)")

  # 3. Build segment description
  companies = [l["display_name"] for l in leads[:5]]
  titles = list({c["title"] for l in leads for c in l.get("contacts", []) if c.get("title")})[:5]
  segment_desc = (
      f"Segment: '{target['name']}'\nCompanies: {', '.join(companies)}\n"
      f"Titles: {', '.join(titles)}\nCount: {len(leads)}"
  )

  # 4. Generate outreach content
  gen = requests.post(
      "https://app.mavera.io/api/v1/generations",
      headers={"Authorization": f"Bearer {MAVERA_KEY}"},
      json={"prompt": (
          f"Write a cold email for this B2B segment. Under 150 words, soft CTA.\n\n"
          f"{segment_desc}\n\n"
          f"Write 3 variants: (1) Pain-point led, (2) Social proof, (3) Curiosity/question led"
      )},
  )
  gen.raise_for_status()
  print(f"\n{'='*50}\nGenerated content for: {target['name']}\n{'='*50}")
  print(gen.json().get("content", gen.json().get("text", "")))

  # 5. LinkedIn variant
  li = requests.post(
      "https://app.mavera.io/api/v1/generations",
      headers={"Authorization": f"Bearer {MAVERA_KEY}"},
      json={"prompt": f"Write a LinkedIn connection request (under 300 chars) for this segment.\n\n{segment_desc}"},
  )
  li.raise_for_status()
  print(f"\nLinkedIn message:\n{li.json().get('content', li.json().get('text', ''))}")
  ```

  ```javascript JavaScript theme={"dark"}
  const CLOSE_KEY = process.env.CLOSE_API_KEY;
  const CLOSE_AUTH = "Basic " + Buffer.from(`${CLOSE_KEY}:`).toString("base64");
  const MAVERA_KEY = process.env.MAVERA_API_KEY;

  async function closeGet(path, params = {}) {
    const url = new URL(`https://api.close.com/api/v1${path}`);
    for (const [k, v] of Object.entries(params)) url.searchParams.set(k, v);
    const res = await fetch(url, { headers: { Authorization: CLOSE_AUTH } });
    if (!res.ok) throw new Error(`Close ${res.status}`);
    return res.json();
  }

  // 1. List Smart Views
  const views = (await closeGet("/saved_search/", { _type: "lead" })).data || [];
  console.log("Smart Views:", views.map((v) => v.name).join(", "));

  const target = views[0];
  if (!target) { console.log("No views."); process.exit(0); }

  // 2. Fetch matching leads
  const leads = (await closeGet("/lead/", {
    query: target.query || "", _limit: 50, _fields: "id,display_name,contacts,custom",
  })).data || [];
  console.log(`\nUsing: ${target.name} (${leads.length} leads)`);

  // 3. Build segment description
  const companies = leads.slice(0, 5).map((l) => l.display_name);
  const titles = [...new Set(leads.flatMap((l) => (l.contacts || []).map((c) => c.title).filter(Boolean)))].slice(0, 5);
  const segmentDesc = `Segment: '${target.name}'\nCompanies: ${companies.join(", ")}\nTitles: ${titles.join(", ")}\nCount: ${leads.length}`;

  // 4. Generate content
  const genRes = await fetch("https://app.mavera.io/api/v1/generations", {
    method: "POST",
    headers: { Authorization: `Bearer ${MAVERA_KEY}`, "Content-Type": "application/json" },
    body: JSON.stringify({ prompt:
      `Write a cold email for this B2B segment. Under 150 words, soft CTA.\n\n${segmentDesc}\n\n` +
      `3 variants: (1) Pain-point, (2) Social proof, (3) Curiosity` }),
  });
  const gen = await genRes.json();
  console.log(`\n${"=".repeat(50)}\nGenerated for: ${target.name}\n${"=".repeat(50)}`);
  console.log(gen.content || gen.text || "");

  // 5. LinkedIn variant
  const liRes = await fetch("https://app.mavera.io/api/v1/generations", {
    method: "POST",
    headers: { Authorization: `Bearer ${MAVERA_KEY}`, "Content-Type": "application/json" },
    body: JSON.stringify({ prompt: `LinkedIn connection request (under 300 chars) for:\n\n${segmentDesc}` }),
  });
  console.log(`\nLinkedIn:\n${(await liRes.json()).content || ""}`);
  ```
</CodeGroup>

## Example Output

```text theme={"dark"}
Using: Enterprise FinTech (34 leads)

**Variant 1 — Pain-point led:**
Subject: Your compliance team is spending 40 hours/week on reports
Hi {{first_name}}, FinTech compliance teams are drowning in manual
reporting. Our platform automates SOX and SOC 2 reports, cutting
prep time by 80%. Worth a 15-minute walkthrough this week?

**Variant 2 — Social proof:**
Subject: How Acme Payments cut audit prep from 3 weeks to 2 days
Hi {{first_name}}, Acme Payments reduced audit prep from 3 weeks
to 2 days. Happy to share the case study if useful.

**Variant 3 — Curiosity:**
Subject: Quick question about {{company}}'s reporting stack
Hi {{first_name}}, curious — still using spreadsheets for regulatory
reporting? I have a benchmark report for FinTech your size.

LinkedIn: "Hi {{first_name}} — I work with FinTech compliance
teams on automating regulatory reporting. Would love to connect."
```

## Error Handling

| Error                      | Cause                     | Fix                                               |
| -------------------------- | ------------------------- | ------------------------------------------------- |
| Smart View returns 0 leads | Criteria too narrow       | Verify in Close UI; update filters                |
| `query` field empty        | Internal format           | Use `GET /lead/?saved_search_id={id}` instead     |
| Generation too generic     | Segment lacks specificity | Enrich with custom fields or recent activity data |
| Rate limit on generations  | Concurrent calls          | Queue with 500ms delay between batches            |