Cline ranked #1 among 100 evaluated agents with a score of 87.8 in the week of 2026-10-12, according to the weekly agent rankings. Deepseek-harness posted the largest drop, falling from rank 16 to ran…
A developer at Assay (伊洛科技有限公司) built a 30-question exam to test whether AI agents can recall their own operational history, and found their own grader was wrong on 7 of 18 items, including two incorr…
A developer benchmarked six deduplication-and-update strategies for agent long-term memory systems, testing mem0 2.0.7, MemOS and signetai on the open-source NeatMem project. In a reproduction using O…
A developer working on the Lakekeeper catalog built a runnable example extending zero trust to agent memory, in which two agents keep private memory in a single catalog, propose skills they cannot pub…
A first conformance census of the Testimony Record specification found that none of eight agent memory and framework systems—OMEM, mem0, Graphiti, LangGraph, CrewAI, the OpenAI Agents SDK, AutoGen, an…
AWS Strands Agents Tools, the first-party tool package for the Strands Agents SDK, received four CVEs between July 15 and August 6, 2026, all stemming from security-sensitive parameters exposed as LLM…
DeepSeek's deepseek-harness agent framework reached 144,024 GitHub stars within four days of release, debuting at No. 28 in the Agent Arena weekly ranking for the week of 2026-08-17. The ranking, whic…
A developer's evaluation of three agent memory frameworks—file-based, structured store, and reinforcement-learning-trained experience—found that the file-based approach, as implemented in OpenClaw's m…
A four-way comparison of Claude memory tools finds that LoreConvo, a local-first SQLite-based memory layer built by the author, offers guaranteed privacy and cross-client support via MCP, while the Cl…
A new benchmark study from a40-labs measures three shapes of agent memory—file-based curation, structured stores, and trained experience—and finds that the file-based approach used by Claude Code, Cli…
Semvec, a semantic memory system from Versino PsiOmega, achieves near-mem0 quality on the LOCOMO benchmark with zero generative-LLM calls at ingest, reducing context tokens by 87% and ingest time by ~…
Mem0, an open-source universal memory layer for AI agents, shipped an update introducing single-pass ADD-only extraction, agent-generated facts as first-class data, entity linking, multi-signal retrie…
A 2026 guide covers Microsoft's GraphRAG, an open-source graph-based retrieval-augmented generation system with 33K GitHub stars, alongside other tools like AnythingLLM, RAGFlow, and Arize AI Phoenix …
A new website, Curated Collections, offers scenario-driven tool stacks that assemble 5-10 deep-dive tools into complete solutions for specific user scenarios, including AI agent ethics, multi-agent wo…
OpenClaw and Hermes-Agent developers have four memory options for action-taking agents: plain markdown files, Obsidian, mem0, and local vector memory. The choice depends on whether the agent needs sim…
Enki Labs released evaluation results for its closed-source memory engine Enki, showing comparable answer accuracy to mem0 on the LongMemEval-S benchmark while storing roughly half the facts (138 vs 2…
CtxGov, a new open-source developer tool, launched v0.9.0 to diagnose and govern AI agent context and memory-state influence before execution. The tool provides local-first CLI commands for change-gat…
Kage, an open-source framework for AI coding agents, introduces verified memory that is stored as files in a git repository, with citations checked against the codebase before writing and stale entrie…
Agent Memory Guard, an OWASP Incubator Project, has been released as a runtime defense layer that screens all reads and writes to AI agent memory to block prompt injection, secret leakage, and integri…
An engineering manager with two teams of 14 engineers reported that individual AI memory systems create a critical knowledge gap in team workflows, where one engineer's AI learns from debugging a Stri…