Tencent Cloud Open-Sources TencentDB Agent Memory v2.0, a Team-Level Memory Hub for AI Coding Agents

Tencent Cloud has open-sourced TencentDB Agent Memory v2.0, a team-level memory hub for AI agents, designed to avoid repeating project context across sessions. The system turns conversations, documents, and code into four reusable memory assets—Chat Memory, Skill, LLM-Wiki, and CodeGraph—that are versioned, permissioned, and equipped to specific agents. What is new is the governance layer that allows team agents to share learned context while keeping private items secure. The stable 2.0.0 release was published on August 3, 2026. It is MIT-licensed and self-hosted, with Docker images on Docker Hub supporting linux/amd64 and linux/arm64. The project targets solo builders and small engineering teams, with mid-size organizations able to run it as shared infrastructure, while large regulated enterprises should pilot it. Chat Memory is distilled into layers L0, L1, L2, and L3, with retrieval using BM25, vector search, and RRF under budget caps. Governance uses visibility levels (private, team, restricted) and fixed binding plus ACL. The Memory Proxy supports Anthropic and OpenAI protocols, and official SDKs are available for TypeScript and Python. Supported integrations include OpenClaw, Hermes, Claude Code, CodeBuddy, and direct SDK use. Reported results claim PersonaMem accuracy rose from 48% to 76%, a 59% relative improvement, but this is self-reported and not independently reproduced.