Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.chatgrid.ai/llms.txt

Use this file to discover all available pages before exploring further.

What you’ll build

An automated client onboarding workflow that creates a structured knowledge board from a client’s public web presence and industry data.
  • Board pre-populated with client website and industry reports
  • AI-generated industry overview, stakeholder analysis, and key challenges
  • Structured zones with action-item sticky notes
  • Ready to share with the consulting or agency team

Prerequisites

  • ChatGrid API key (get one here)
  • Node.js 18+ or Python 3.10+

Step 1: Create a client board

curl -X POST https://api.chatgrid.ai/v1/boards \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"name": "Client: Meridian Health Systems - Onboarding"}'

Step 2: Add client website and industry report

Batch-add the client’s public pages and a relevant industry report as source nodes.
curl -X POST https://api.chatgrid.ai/v1/boards/{boardId}/nodes/batch \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "nodes": [
      {"type": "website", "data": {"url": "https://meridianhealth.com/about"}},
      {"type": "website", "data": {"url": "https://meridianhealth.com/leadership"}},
      {"type": "website", "data": {"url": "https://deloitte.com/insights/healthcare-outlook-2026"}}
    ]
  }'

Step 3: Vectorize all sources

for url in \
  "https://meridianhealth.com/about" \
  "https://meridianhealth.com/leadership" \
  "https://deloitte.com/insights/healthcare-outlook-2026"; do
  curl -X POST https://api.chatgrid.ai/v1/boards/{boardId}/documents/vectorize \
    -H "Authorization: Bearer YOUR_API_KEY" \
    -H "Content-Type: application/json" \
    -d "{\"url\": \"$url\"}"
done

Step 4: Generate AI analysis

Create a chat and ask AI for three deliverables: industry overview, stakeholder analysis, and key challenges.
curl -X POST https://api.chatgrid.ai/v1/boards/{boardId}/chats \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"name": "Onboarding Analysis"}'

curl -X POST https://api.chatgrid.ai/v1/boards/{boardId}/chats/{chatId}/messages \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "content": "Based on the client materials and industry report on this board, generate:\n1. Industry Overview: trends, market dynamics, regulatory landscape\n2. Stakeholder Analysis: key executives and their likely priorities\n3. Key Challenges: top 5 challenges this client faces\nBe specific to this client, not generic.",
    "stream": false
  }'

Step 5: Create structured zones with batch nodes

Organize the board into clear zones: one notepad per section, plus action-item sticky notes.
curl -X POST https://api.chatgrid.ai/v1/boards/{boardId}/nodes/batch \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "nodes": [
      {"type": "notepad", "data": {"content": "# Industry Overview\n\n[Paste section 1 from AI output]"}},
      {"type": "notepad", "data": {"content": "# Stakeholder Analysis\n\n[Paste section 2 from AI output]"}},
      {"type": "notepad", "data": {"content": "# Key Challenges\n\n[Paste section 3 from AI output]"}},
      {"type": "sticky_note", "data": {"content": "ACTION: Schedule discovery call with CTO to validate tech priorities"}},
      {"type": "sticky_note", "data": {"content": "ACTION: Request org chart from client contact"}},
      {"type": "sticky_note", "data": {"content": "ACTION: Prepare pitch deck addressing top 3 challenges"}}
    ]
  }'

What’s happening under the hood

ChatGrid fetches each URL, extracts readable content (stripping navigation and boilerplate), and stores it as a document. Vectorization chunks documents into ~500-token segments and embeds them with pgvector. When the AI generates the analysis, it searches across all vectorized content to find relevant passages from both the client’s website and the industry report, producing a response grounded in real data rather than generic knowledge. The batch node endpoint creates all zone nodes atomically in a single transaction.

Next steps

Research Pipeline

Go deeper with multi-source research synthesis

Team Knowledge

Share the onboarding board with your whole team