Overview Research Convenings Prototypes Writings Talks People
Newsletter

Persistent memory is becoming a core layer of agent infrastructure faster than shared exchange semantics are emerging

Every major coding-agent harness (Claude Code, Codex, goose, OpenClaw, Letta Code) implements memory, but each uses a different convention. A developer cannot move a stateful agent from one harness to another. An enterprise cannot switch memory providers without re-ingesting derived memory objects. Open-source memory projects duplicate one another because there is no shared object model, provenance format, or exchange protocol to converge on.

The Open Memory Protocol (OMPI) closes this gap. The Open Memory Protocol Initiative (OMPI) is the working group hosted at the AI Disclosures Project that develops, maintains, and grows the ecosystem around it.

The smallest interoperable layer over existing memory primitives

Backends, ownership, and internal architecture stay with implementers

What is available today

Draft specification

Agent Memory Specification, v0.1

Draft authored by Charles Packer (CEO, Letta) with feedback from the AI Disclosures Project team. Circulating in the working group in advance of the September 9 kickoff.

Read the draft
Reference implementation

Python package (pre-alpha)

A minimal Python implementation of the loader, validator, and harness contract. Runs against any spec-conformant memory directory.

View on GitHub
Governance

Working-group governance model

Steering committee, RFC process, licensing, and long-term stewardship plan (target: standards-body handoff after production adoption).

Read governance
Ecosystem scoping

State of agent memory (2026)

A survey of production and open-source memory systems (Letta, Mem0, Zep, Graphiti, AWS AgentCore Memory, Google OKF, and coding-agent harnesses) and where OMPI fits.

Read the survey

Formal in-kind partners and implementation collaborators

In-kind partner

Mozilla

Co-hosts working-group convenings and leads ecosystem recruitment across Mozilla's open-source and AI-developer network.

Raffi Krikorian (CTO) · Matt Harris
In-kind partner

IBM

Co-hosts convenings, contributes enterprise-adoption pathways, and advises on standards-body process.

Gabe Goodhart (Chief Architect) · Anna Lisa Gentile · Rynne Whitnah
Specification collaborator

Letta

Author of the draft OMPI specification. Contributes MemGPT / Letta memory-object model and the open-source Trajectory package.

Charles Packer (Co-founder and CEO)
Implementation collaborator

Block / goose

Open-source agent harness built on the Model Context Protocol. First-vertical implementation target for OMPI conformance.

goose maintainer team

From kickoff to IETF handoff

Sept 9, 2026
Kickoff technical convening (online). Co-hosted by AI Disclosures Project, Mozilla, and IBM. Working-group participants update the group on their memory work and identify areas for cross-stack coordination.
Oct 2026
In-person session at the O'Reilly open-source unconference (Berkeley). Reference-implementation v0.1 published and installable.
Year 1 (2026–27)
Two coding-agent harnesses shipping conformant memory export/import. First cloud-provider technical spike. Quarterly public adoption reports.
Year 2 (2027–28)
Written specification v1.0. Six coding-agent harnesses conformant. End-to-end benchmark passing across ten harness pairs. Preliminary IETF engagement (Birds-of-a-Feather session or working-group formation).
End of award
IETF Internet-Draft submitted. Long-term stewardship handed to standards-body home (target: Agentic AI Foundation).

Join the working group

If you build agent memory, maintain a coding-agent harness, run a memory-layer service, or work on standards for AI infrastructure, we want you in the working group.

Email OMPI GitHub