Bring the same memory to every AI agent.

Dijin gives Claude, ChatGPT, Cursor, Zed, and future agents a shared memory layer — owned by the user, scoped by permission, and backed by evidence.

Join the developer waitlist Read the protocol

Lock B partner status

Every answer Dijin returns to a partner client is cryptographically signed. Below is which clients are at which Phase of the partner trust chain.

Client	Vendor	Phase	Last verified	Transport
Claude Desktop	Anthropic	Phase 1, Signed Footer	2026-05-28	MCP_SSE
ChatGPT (Desktop)	OpenAI	Phase 1, Signed Footer	2026-05-28	MCP_SSE
Cursor	Anysphere	Phase 1, Signed Footer	2026-05-28	MCP_SSE
Zed	Zed Industries	Phase 1, Signed Footer	2026-05-28	MCP_SSE
Continue	Continue.dev	Phase 1, Signed Footer	2026-05-28	MCP_SSE
Gemini (CLI)	Google	Pending	Not verified yet	MCP_SSE

Partner clients are listed only after at least one successful evidence-bearing round-trip against mcp.dijin.co/mcp, no aspirational logos.

What your AI can ask

13 tools available over the Model Context Protocol (MCP).

searchMemory

agents.tools.searchMemory

agents.tools.searchMemoryDesc

listEntities

List people and topics

See who and what your memory mentions.

getDailySummary

Daily summary

A concise recap of any day.

getWeeklyReview

Weekly review

What stood out across the week.

listActionItems

Action items

Tasks and follow-ups your AI can act on.

getDecisions

Decisions made

What was decided, when, and by whom.

getConflicts

Conflicts and contradictions

Surface where memory disagrees with itself.

getAuditLog

Audit log

Every AI access is logged for you to review.

whoami

Identity

Confirms which account the AI is connected to.

Five lines to connect Claude.

Add this to your Claude Desktop config and Dijin becomes a memory provider.

Save to: ~/Library/Application Support/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "Dijin": {
      "url": "https://mcp.dijin.co/mcp",
      "transport": "sse"
    }
  }
}

Why agents need a memory layer

Every AI agent has a different view of your work. Claude may know the discussion. Cursor may know the code. GitHub may know the commit. Teams may know the meeting. None of them should become the owner of your memory.

one memory the user owns
one permission model
one evidence trail
one audit log
one portable DMF export

Agent-to-agent memory, without agent-to-agent leakage

Dijin does not make agents pass private context directly to each other.

Agent A proposes memory to Dijin.

Dijin verifies, scopes, and stores it with evidence.

Agent B recalls the approved memory later.

The user can inspect, revoke, or export everything.

Why Dijin and not a vector database?

Open standard, not a black box

DMF is published, versioned and portable. Your AI can switch tools without losing memory.

Evidence on every answer

Each result links back to the source moment, your AI can show its work.

You stay in charge

Every connection is consent-based. Every access is logged. Revoke at any time.

Bring the same memory to every AI agent.

Join the developer waitlist Read the protocol