Memory & contextAgent frameworks

TencentDB Agent Memory

TencentCloud/TencentDB-Agent-Memory

Team agent memory hub — Chat Memory, Skills, and LLM-Wiki shared across agents.

Repository
Stars
4.4k
Forks
600
Open issues
30
Language
Go
Licence
MIT
Created
2026-05-10
Last push
2026-08-16
Effort to adopt

Engineering project

Weeks, not days. Treat it as infrastructure with an owner and a budget.

01

What it actually does

A massive memory hub for enterprise agent networks. Instead of each AI agent having amnesia when a session ends, TencentDB Agent Memory stores conversational history, learned skills, and a shared 'LLM-Wiki' in a central database (powered by TencentDB). If Agent A figures out how to fix a complex deployment bug on Monday, it stores that knowledge in the hub. When Agent B encounters a similar bug on Thursday, it queries the memory hub and applies the fix instantly.

02

Who it's for

  • 01

    Large enterprise teams deploying swarms of autonomous agents that need to share knowledge

  • 02

    Customer support teams where hand-offs between different AI bots require perfect context retention

  • 03

    Founders wanting to build an institutional 'AI Brain' that gets smarter as agents do more work

OperationsEngineeringSaaSFintechAny industry
03

Where it earns its keep

  • Creating a shared 'engineering brain' where every bug fixed by an AI agent is documented for future agents
  • Running a 24/7 customer support swarm where agents read past interactions to personalize their responses
  • Building a corporate AI wiki that agents update autonomously based on Slack conversations
04

Use it, or skip it

Reach for it when

  • You have multiple agents working simultaneously and they need to share state and knowledge
  • You want your AI agents to have long-term episodic memory across months of interactions

Skip it when

  • You just use ChatGPT in the browser—this is for heavy enterprise API integrations
  • You don't want to manage a complex TencentDB or PostgreSQL infrastructure
05

10 automations

  1. 01Engineering

    Shared Bug Resolution Brain

    Every time a debugging agent fixes an issue, it writes the root cause and solution to the hub so other agents can fix it instantly next time.

  2. 02Operations

    Personalized Support Swarm

    Support agents query the memory hub to remember a user's frustration level from three months ago, adapting their tone accordingly.

  3. 03Operations

    Automated Onboarding Wiki

    Agents monitor Slack for answers to common new-hire questions and compile them into a shared, agent-readable wiki.

  4. 04Marketing

    Cross-Department Insight Sharing

    Marketing agents read summaries generated by Sales agents in the memory hub to adjust ad copy based on real customer objections.

  5. 05Founders

    Persistent Executive Assistant

    An AI assistant that remembers your preferences, meeting habits, and project context across years of interactions.

  6. 06Engineering

    Agent Skill Marketplace

    When an agent successfully writes a new Python script to solve a problem, it uploads the 'skill' to the hub for other agents to execute.

  7. 07Sales

    Customer Lifecycle Tracker

    Track the entire lifecycle of a customer across multiple agent interactions (sales, support, success) in a unified memory timeline.

  8. 08Engineering

    Automated Post-Mortem Generator

    After an outage, an agent queries the memory hub to reconstruct the exact timeline of what other agents did during the incident.

  9. 09Engineering

    Contextual Code Reviewer

    A PR agent checks the memory hub for past PR feedback to ensure developers aren't repeating the same architectural mistakes.

  10. 10Founders

    Dynamic Product Roadmap

    Product agents aggregate thousands of feature requests from support agents in the hub to prioritize the engineering backlog.

Want one of these running by Friday?

LimeDock builds these as real workflows inside your stack — deployed to your cloud, wired into your Slack and CRM, with the code in your repo. You pay a build fee and your own API keys, nothing else.

Book a workflow call
07

Source

Repository stats were read from the GitHub API and reflect the last time we refreshed this entry. The editorial breakdown above is LimeDock’s own analysis — we are not affiliated with TencentCloud.

https://github.com/TencentCloud/TencentDB-Agent-Memory