Resources Overview

What is a Resource?

A Resource in Vaillancore is a collection of knowledge or reference material that an AI agent can pull from when generating responses. It's a way to give your agents specialized knowledge — documentation, FAQs, policies, SOPs, product details, or anything else your AI should be able to understand and respond with.

You manage Resources entirely through the Admin Dashboard — no coding or API usage required.

How It Works

When you create a Resource, you:

  • Give it a name and description (for your own reference)
  • Add one or more Embeddings — these are individual content blocks

Each Embedding becomes a vectorized chunk of knowledge (an embedding), which is searchable and ranked by relevance when a user sends a message. This is how your agent "remembers" relevant parts of your uploaded information.

Embeddings: The Core of a Resource

An Embedding in the dashboard is a single block of content, like a paragraph or small section. Each one is individually indexed and retrieved based on semantic similarity (not just keyword matching).

Best Practices for Embedding Size

  • Keep it between 1–3 paragraphs or 500–750 words max
  • Avoid large, dense content blocks — split them into logical pieces
  • Think: one topic or idea per embedding

✅ Good: "Our return policy allows customers to request a refund within 30 days, provided the item is unused and in original packaging."

❌ Bad: An entire 5-page refund policy pasted into one field

The more precise and segmented your content blocks are, the better the AI can retrieve and respond with relevance.

Connecting Resources to Agents

Once you've created a Resource, you can attach it to any of your AI agents from the Agent Edit screen in the Admin Dashboard:

  1. Go to your Agents tab
  2. Click an agent
  3. Under Attached Resources, select one or more Resources
  4. Save changes

Now, whenever a message is sent to that agent, Vaillancore automatically pulls the most relevant embeddings from those Resources and includes them in the AI's context — no manual retrieval or search needed.

Example Use Cases

Resource NameUsed For
"Internal HR Guide"Onboarding bot for new employees
"Product FAQ"Support assistant answering user queries
"Security Protocols"AI agent handling IT or compliance chats

Pro Tips

  • Break large documents into multiple logical Resources (e.g., "Returns Policy", "Shipping Info", etc.)
  • Use consistent formatting and tone within each embedding
  • Keep Embeddings focused — don't blend multiple topics into one