Knowledge Base

The Knowledge Base lets you upload documents that your AI agent references when answering questions. This turns a general-purpose AI into a domain expert for your community.

How It Works

When a community member asks a question, the agent:

Searches the knowledge base for relevant passages (semantic search)
Uses matching content as context for generating a response
Responds with information grounded in your documents

This means the agent can accurately answer questions about your project, product, or community — not just general knowledge.

Capacity Limits

Dimension	Limit
Documents per Bot	50
Body size per document	50 KB (~50,000 English characters)
Title length	256 bytes

Split larger sources into multiple documents, or remove older ones before adding new content.

Adding Documents

Navigate to Knowledge

Open Knowledge from the sidebar.

Create a Document

Click Add Document and fill in:

Title — a descriptive name (e.g. "FAQ", "Tokenomics", "Community Rules")
Summary (optional) — a 1–2 sentence overview for internal management
Content — paste or type the document body (Markdown or plain text)

Save

Click Save. The document is semantically indexed and available to the Bot as soon as indexing completes.

Best Practices

Keep documents focused. One topic per document produces better search results than a single large document covering everything.

Good use cases:

Frequently Asked Questions (FAQ)
Product or project documentation
Community rules and guidelines
Support troubleshooting guides
Onboarding information for new members

Tips for quality:

Write in clear, concise language
Include common variations of questions users might ask
Update documents when information changes
Remove outdated content to avoid contradictions

Documents are processed and indexed for semantic search. The agent finds relevant content even when users don't use exact keywords from your documents.