Governed memory you can trust.

The authoritative source of your organisation's shared AI memories.

Apply for Beta Read the docsD

Lore Index dashboard showing the entry registry under review

Store your Lore in an authoritative index, curated and approved for your team, by your team.

Each piece of Lore represents your internal processes, glossaries, contacts, and guidelines.

Fight hallucinations with verified records of trust.

Agent

I have some questions about our CI/CD, who do I ask?

Called Lore

John Smith builds and maintains all the CI/CD workflows — reach out to him.

Fight context bloat by giving the agent only what it needs, when it needs it.

Agent

Called Lore

Bootstrap pack loaded. The pack returned 12 MUST instructions, plus a tag index. ~750 tokens, about 0.375% of context.

Audit every action.

All Lore retains its full edit history, to be compared anytime.

The Lore Index audit log recording every action taken on the registry

Version history comparing two approved revisions of a Lore entry

Scope by team and role.

Create authoritative knowledge bases for every team. Each team only sees their Lore, keeping context bloat at bay.

The teams screen where each team sees only its own scoped Lore

An easy-to-use MCP for any agent.

Point your agent at one endpoint. It loads the bootstrap pack at the start of every session, then searches the index for anything more it needs.

Compatible with all MCP-enabled agents
Task-aware search keeps agent context light and relevant

~/.config/mcp/lore.json

// Add Lore Index to any MCP-capable agent
{
  "mcpServers": {
    "lore": {
      "url": "https://api.loreindex.ca/mcp",
      "headers": {
        "Authorization": "Bearer lore_sk_..."
      }
    }
  }
}

→ bootstrap loaded · 1,480 tokens · 12 MUST · 34 SHOULD

How to index your Lore.

01
Write a new Lore.
02
Each Lore gets a title, description, category, scope, and status.
03
Scope the new Lore to the appropriate team or role.
04
Edit Lore and review past versions.
05
Agents query the Lore when they need it.

A Lore entry showing its title, scope, category, and approved body

Agent

Create a CRM website for internal use

Called Lore

Creating a plan to create an internal CRM following your org's preferred technical stack: SvelteKit, BetterAuth, and PostgreSQL.

Agent

Create a CRM website for internal use

Called Lore

Creating a plan to create an internal CRM following your org's preferred technical stack: SvelteKit, BetterAuth, and PostgreSQL.

Extraction guesses. Governance decides.

“Why doesn’t Lore Index track these changes for me and have AI write the Lore?”

Lore isn’t controlled or documented by AI, because AI doesn’t know your organisation like you do. The Index is the single source of truth. Every piece of Lore must be approved by an Index Administrator, whether it was drafted by an employee or AI. Agents with the proposal scope can suggest new Lore or edits to existing Lore, but cannot publish new Lore by default.

This strict layer of trust ensures nothing irrelevant or incorrect is passed to agents and treated as gospel. It keeps knowledge outside of an agent’s vendor and in a dataset you control. Context is kept clean and lean, and every entry becomes a binding record.

FounderGoverned memory you can trust.

Make your organisation’s memory a matter of record.

Apply for Beta See pricing