Governance, not extraction

Organizational memory your agents can actually be trusted with.

The instructions, preferences, facts, and glossary new hires learn by osmosis — written once, curated by humans, scoped by team, and served to any agent over one MCP endpoint.

Create your workspaceRead the docs
◆ Human-approved◆ Vendor-neutral◆ ~1,500 tokens
how memory reaches your agents● live
Your team writes it down
MUST rulesSHOULDGlossaryContacts
CURATE · APPROVE · VERSION↓
Lore IndexGOVERNED CORE
scopedversionedaudited
Compiled bootstrap pack · caps context at ~1,500 tokens
ONE MCP ENDPOINT↓
Claude
via MCP
Cursor
via MCP
Your agent
via MCP
Trusted by teams who take governance seriously
NorthwindHalden LabsMeridianKestrel AIVerto
01

Governed

Every entry is human-curated, scoped, versioned, and audited. Agents propose; people approve. MUST rules are binding — not scraped guesses.

02

Vendor-neutral

One MCP endpoint any agent can speak to — Claude, Cursor, your own. No lock-in, no per-vendor memory silos. Self-host it or run it in the cloud.

03

Token-frugal

A compiled bootstrap pack caps context at ~1,500 tokens: the org's active MUST/SHOULD guidance, narrow-scope-first. Load once, search narrowly.

How it works

Write it down once. Everything else follows.

1

Write

Capture rules, preferences, facts, and glossary as structured entries — in plain language.

2

Scope

Assign each entry to a team, project, or the whole org. Narrow scope wins first.

3

Version

Humans approve changes. Every edit is versioned and audited — you can always see who set what.

4

Serve

Agents bootstrap the pack at session start and cite the rules they followed.

One endpoint

Point any agent at Lore Index and it just knows your team.

Add the MCP server once. At the start of every session your agent pulls the compiled bootstrap pack — active MUST/SHOULD guidance, narrowest scope first — and cites what it followed.

  • ◆ No per-vendor SDKs — plain MCP
  • ◆ Search narrows to the task at hand
  • ◆ Self-host, or run it in our cloud
~/.config/mcp/lore.json
// Add Lore Index to any MCP-capable agent
{
  "mcpServers": {
    "lore": {
      "url": "https://api.loreindex.ca/mcp",
      "headers": {
        "Authorization": "Bearer compass_sk_..."
      }
    }
  }
}
→ bootstrap loaded · 1,480 tokens · 12 MUST · 34 SHOULD
The difference

Extraction guesses. Governance decides.

Memory by extraction
  • ✕ Scraped from chats — nobody approved it
  • ✕ Silos per vendor, drifts out of date
  • ✕ Balloons context, hard to audit
  • ✕ No way to say "this is binding"
Memory by governance
  • ✓ Human-curated, approved, and cited
  • ✓ One vendor-neutral MCP endpoint
  • ✓ ~1,500-token compiled bootstrap
  • ✓ MUST rules are binding by design

Stop letting every agent relearn how your team works.

Write it down once. Scope it. Version it. Let agents bootstrap it at the start of every session.

Create your workspaceSee pricing