AMP: A Vendor-Neutral Wire Format for Agent Memory Operations

Computer Science > Cryptography and Security [Submitted on 31 May 2026] AMP: A Vendor-Neutral Wire Format for Agent Memory Operations Thamilvendhan Munirathinam Agent-memory frameworks - mem0, Letta/MemGPT, Cognee, Zep/Graphiti, MemoryOS, MemTensor - each ship their own SDK, storage layout, and operational vocabulary. There is no shared wire format: every integration is bespoke, every migration rebuilds memory from scratch, and no framework ships a governance surface that lets a human review writes before they enter long-term storage. We present memorywire, a JSON-Schema 2020-12 wire format for five memory operations (remember, recall, forget, merge, expire) over four memory types (semantic, episodic, procedural, emotional), with a MemoryStore interface, a fan-out router, and an optional HITL governance channel. We describe an open-source reference implementation with five backend adapters (sqlite-vec, mem0, Letta, Cognee, pgvector); a microbenchmark on a 100-fact / 50-query labelled corpus achieving recall@5 = 1.000 on the 42 labelled queries with ingest p50 = 37.8 ms and recall p50 = 40.6 ms; an adversarial-fusion experiment showing Reciprocal Rank Fusion holds recall@5 = 1.000 across a 1-of-N rank-0 injection sweep (K in {0,5,...,50}) where max fusion collapses to 0.500 with 80% leak at K >= 5; and a 16-scenario cross-adapter conformance suite passing 68 of 80 cells with zero failures. The contribution is not a new algorithm; it is a packaging of established components (RRF, FSMs, STM/LTM consolidation, diff-and-approve workflows) into a venue-neutral protocol with an empirically validated reference, positioned to compose with the Model Context Protocol rather than compete with it. Comments: 17 pages, 1 figure, 6 tables. Reference implementation with 5 backend adapters (sqlite-vec, mem0, Letta, Cognee, pgvector), governance UI, microbench, adversarial-fusion experiment, 16-scenario conformance suite, threat model, and preliminary LongMemEval + LoCoMo numbers. Code at this https URL. Companion to arXiv:2604.18248 (Prompt Injection Detection) Subjects: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Distributed, Parallel, and Cluster Computing (cs.DC) Cite as: arXiv:2606.01138 [cs.CR]   (or arXiv:2606.01138v1 [cs.CR] for this version)   https://doi.org/10.48550/arXiv.2606.01138 Focus to learn more Submission history From: Thamilvendhan Munirathinam [view email] [v1] Sun, 31 May 2026 10:18:56 UTC (80 KB) Access Paper: HTML (experimental) view license Current browse context: cs.CR < prev   |   next > new | recent | 2026-06 Change to browse by: cs cs.AI cs.DC References & Citations NASA ADS Google Scholar Semantic Scholar Export BibTeX Citation Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer (What is the Explorer?) Connected Papers Toggle Connected Papers (What is Connected Papers?) Litmaps Toggle Litmaps (What is Litmaps?) scite.ai Toggle scite Smart Citations (What are Smart Citations?) Code, Data, Media Demos Related Papers About arXivLabs Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)