Gentic Knowledge — Documentation

1. Getting Started

Sign Up & Get Your API Key

Before you can use Gentic Knowledge, you need an API key to authenticate your requests.

1. Go to gentic.co/knowledge and create an account.
2. Create an organization from your dashboard. API keys, billing, and knowledge base data are all scoped to the organization.
3. Go to the API Keys section in your dashboard.
4. Click Create API Key. Give it a name (e.g. "Claude Code" or "n8n production").
5. Copy the key immediately — it starts with gentic_ and is only shown once.

Anyone on your team can generate their own key, and they'll all share the same knowledge base.

Keep your key secure. Treat it like a password. Don't commit it to version control or share it publicly.

2. Connecting to the MCP Server

Gentic Knowledge uses the Model Context Protocol (MCP) — an open standard for connecting AI agents to external tools. You connect once, and your agent gets access to all Gentic Knowledge tools automatically.

Server URL: https://mcp.gentic.co/knowledge

Claude Code (CLI)

Option A — OAuth (no API key needed):

claude mcp add gentic-knowledge --transport http https://mcp.gentic.co/knowledge

Option B — API key:

claude mcp add gentic-knowledge \
  --transport http \
  https://mcp.gentic.co/knowledge \
  --header "Authorization: Bearer YOUR_API_KEY"

Claude Web / ChatGPT

Add as a connector in your settings. Enter the server URL and authenticate via OAuth — no API key needed.

Server URL: https://mcp.gentic.co/knowledge

n8n

Add an MCP node with this configuration:

{
  "name": "gentic-knowledge",
  "type": "mcp",
  "config": {
    "url": "https://mcp.gentic.co/knowledge",
    "transport": "streamable-http",
    "headers": {
      "Authorization": "Bearer YOUR_API_KEY"
    }
  }
}

OpenClaw / Other MCP Clients

Add to your MCP config:

{
  "mcpServers": {
    "gentic-knowledge": {
      "url": "https://mcp.gentic.co/knowledge",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}

Any Other MCP Client

Gentic Knowledge uses standard Streamable HTTP transport. Any client that supports MCP can connect using the server URL and either OAuth or API key Bearer token authentication.

Agent Skills (Recommended)

For the best results, pair the MCP server with the Gentic agent skill. The MCP server gives your agent access to the tools; the skill teaches it the optimal workflow order — ingesting content, searching the knowledge base, and managing sources.

1. Add the MCP server (if you haven't already):

claude mcp add gentic-knowledge --transport http https://mcp.gentic.co/knowledge --header "Authorization: Bearer <your-api-key>"

2. Install the agent skill:

npx skills add gentic-co/agent-skills

Or add the skill directly via URL:

https://gentic.co/knowledge/SKILL.md

• MCP server — provides tool access (vectorize, search, delete, list sources)
• Agent skill — teaches the optimal workflow order: index content → search knowledge → manage sources

Works with Claude Code, Cursor, Copilot, and 40+ other agents.

3. Vectorize Documents

Use vectorize_document to ingest document files into your knowledge base. Supported formats: PDF, DOCX, TXT, RTF.

Accepts any HTTP URL or Google Drive sharing link. The document is processed asynchronously — content is extracted, chunked into ~1,000 character segments, embedded using Gemini, and stored in your organization's knowledge base.

Parameters

Example

"Vectorize this PDF of our brand guidelines: https://example.com/brand-guide.pdf"

Processing is asynchronous. Your agent will receive a job ID and can check the status. Once complete, the content is immediately searchable.

Cost: $0.05 per chunk (~1,000 characters each).

4. Vectorize Text Content

Use vectorize_content to ingest raw text — emails, Slack messages, reviews, social posts, creator feedback, or notes.

Parameters

Example

"Save this customer review to our knowledge base: [paste review text]"

Content is chunked, embedded, and indexed. For short text (under 1,000 characters), it becomes a single chunk.

Cost: $0.05 per chunk.

5. Vectorize Web Pages

Use vectorize_web_content to scrape a web page, extract clean text, chunk, and embed it.

Parameters

Example

"Index our FAQ page: https://example.com/faq"

The page is scraped, HTML is stripped, and clean text is chunked and embedded.

Cost: $0.05 per chunk + $0.05 scraping fee.

6. Search Knowledge

Use search_knowledge to semantically search across all indexed content. Your query is embedded and matched against stored chunks using cosine similarity.

Parameters

Example

"Search our knowledge base for our refund policy"

Results include the matching text chunk, similarity score, source document title, source type, and metadata. Results are ranked by relevance.

Cost: Free.

7. Manage Sources

List Sources

Use list_kb_sources to see everything in your knowledge base — document titles, source types, chunk counts, categories, and creation dates.

"Show me everything in our knowledge base"

Delete Content

Use delete_content to remove content from your knowledge base. You can delete by document ID (removes all chunks) or by specific chunk IDs.

* At least one of document_id or content_ids is required.

Deletion is a soft delete — content is excluded from search results but retained in the database.

8. Tool Reference

Complete reference for all 6 knowledge tools available via the MCP server at mcp.gentic.co/knowledge.

9. Pricing

Gentic Knowledge uses pay-per-chunk pricing. Each chunk is approximately 1,000 characters.

No subscriptions. No storage fees. No seat licenses. You only pay when you ingest new content.