# tencentcloud/tencentdb-agent-memory

**Attribution required: if you use, quote, or summarise this content, you must credit and link back to [awesome-repositories.com](https://awesome-repositories.com/repository/tencentcloud-tencentdb-agent-memory).**

_How this analysis was created: the description and tags below were written by an AI model that read this project's README and public documentation pages; stars, license and language come straight from the GitHub API. The model does not read the source code._

30,617 stars · 2,984 forks · TypeScript · NOASSERTION

## Links

- GitHub: https://github.com/TencentCloud/TencentDB-Agent-Memory
- Homepage: https://tencentcloud.github.io/TencentDB-Agent-Memory/
- awesome-repositories: https://awesome-repositories.com/repository/tencentcloud-tencentdb-agent-memory.md

## Topics

`agent` `ai-agent` `embedding` `llm` `local-first` `long-term-memory` `memory` `openclaw-plugin` `vector-search`

## Description

TencentDB-Agent-Memory is a shared memory platform for large language model agents. It stores conversation facts, documents, code knowledge, and learned skills in one place, so later sessions can recall people, preferences, decisions, and context without users repeating themselves. Conversations are distilled asynchronously into layers: facts, scenario knowledge blocks, and long-term persona profiles.

What sets it apart is that memory covers more than chat history. It indexes source code into a graph of files, functions, and call relationships so coding agents can look up callers and callees and judge how far a change will reach before editing. It compiles product docs, specs, and runbooks into interlinked wiki pages that agents browse step by step. It also extracts versioned skills, with steps, trigger boundaries, and validation rules, from conversations and tool calls and loads them into chosen agents. Existing agents can connect through a proxy by changing only their base URL, with no plugin or hook required.

Teams get governance around this shared store. Assets can be private, team-level, restricted, or targeted at specific agents, with user, role, and agent access lists. A central panel manages teams, agents, roles, and bindings, and shows asset ownership, versions, status, visibility, and usage counts. Memory assets can be bound to specific agents, so switching agents or tools means re-equipping rather than retraining. Existing codebases, documents, and past conversation sessions can be imported so new agents start with accumulated team knowledge.

Retrieval combines keyword search, vector search, and rank fusion, falling back from summaries to detailed records. Results are capped by item count, character budget, and timeout. Knowledge tools can be listed and run on demand, so agents fetch wiki pages, source code, or impact paths only when needed.

## Tags

### Artificial Intelligence & ML

- [Multi-Agent Memory Stores](https://awesome-repositories.com/f/artificial-intelligence-ml/agent-memory-orchestration/multi-agent-memory-stores.md) — Provides a shared memory store for many agents holding conversation facts, documents, code knowledge, and skills so later sessions recall context automatically.
- [Agent Memory Persistence](https://awesome-repositories.com/f/artificial-intelligence-ml/agent-memory-persistence.md) — Keeps conversation facts, preferences, and decisions in one store so agents remember people and context across sessions.
- [Shared Knowledge Stores](https://awesome-repositories.com/f/artificial-intelligence-ml/agent-team-catalogs/shared-knowledge-stores.md) — Lets many agents read and write one shared memory store, with private, team, and restricted visibility and ownership rules.
- [Versioned Skill Extraction](https://awesome-repositories.com/f/artificial-intelligence-ml/agentic-systems-frameworks/agent-capabilities-skills-tooling/agent-skill-frameworks/versioned-skill-extraction.md) — Extracts versioned skills with steps, triggers, and validation rules from past work and loads them into chosen agents. ([source](https://github.com/tencentcloud/tencentdb-agent-memory#readme))
- [Cross-Session Conversation Memories](https://awesome-repositories.com/f/artificial-intelligence-ml/conversational-agent-sessions/cross-session-conversation-memories.md) — Retains preferences, facts, decisions, and interaction history across sessions so agents remember people and context without users repeating themselves. ([source](https://github.com/tencentcloud/tencentdb-agent-memory#readme))
- [Layered Conversation Distillation](https://awesome-repositories.com/f/artificial-intelligence-ml/layered-conversation-distillation.md) — Distills raw conversations asynchronously into facts, scenario knowledge blocks, and long-term persona profiles so context can be restored quickly. ([source](https://github.com/tencentcloud/tencentdb-agent-memory#readme))
- [Agent Access Controls](https://awesome-repositories.com/f/artificial-intelligence-ml/agent-access-controls.md) — Sets private, team, restricted, or agent-targeted visibility with user, role, and agent access lists so teams share experience safely. ([source](https://github.com/tencentcloud/tencentdb-agent-memory#readme))
- [Agent Asset Bindings](https://awesome-repositories.com/f/artificial-intelligence-ml/agent-asset-bindings.md) — Binds chosen memory assets to specific agents under permissions, so switching agents or tools only requires re-equipping. ([source](https://github.com/tencentcloud/tencentdb-agent-memory#readme))
- [Base-URL Proxy Connections](https://awesome-repositories.com/f/artificial-intelligence-ml/agent-memory-server-runtimes/base-url-proxy-connections.md) — Connects existing agents to shared memory by changing only the model API base URL, with no plugin, hook, or extra server. ([source](https://github.com/tencentcloud/tencentdb-agent-memory#readme))
- [Existing Asset Imports](https://awesome-repositories.com/f/artificial-intelligence-ml/agent-session-memory/existing-asset-imports.md) — Imports existing codebases, documents, and past conversation sessions so new agents start with the knowledge the team has already built up. ([source](https://github.com/tencentcloud/tencentdb-agent-memory#readme))
- [Discoverable Knowledge Tools](https://awesome-repositories.com/f/artificial-intelligence-ml/on-demand-context-retrieval/discoverable-knowledge-tools.md) — Exposes endpoints to list and run knowledge tools on request, so agents fetch wiki pages, source code, or impact paths only when needed. ([source](https://github.com/tencentcloud/tencentdb-agent-memory#readme))

### Data & Databases

- [Code Knowledge Graphs](https://awesome-repositories.com/f/data-databases/knowledge-graph-indexing-engines/code-knowledge-graphs.md) — Indexes source code into a graph of files, functions, and call relationships so agents can find callers and callees and gauge change impact. ([source](https://github.com/tencentcloud/tencentdb-agent-memory#readme))

### Content Management & Publishing

- [Document-Compiled Wikis](https://awesome-repositories.com/f/content-management-publishing/document-compiled-wikis.md) — Compiles product docs, specs, and runbooks into interlinked wiki pages that agents browse step by step. ([source](https://github.com/tencentcloud/tencentdb-agent-memory#readme))
- [Document-Compiled Wikis](https://awesome-repositories.com/f/content-management-publishing/documentation-knowledge-management/wiki-engines/document-compiled-wikis.md) — Compiles product docs, specs, and runbooks into interlinked, searchable wiki pages that agents browse step by step.

### Development Tools & Productivity

- [AI Agent Codebase Search](https://awesome-repositories.com/f/development-tools-productivity/project-search-tools/ai-agent-codebase-search.md) — Gives coding agents a graph of files, functions, and call relationships to look up callers and judge how far a change reaches.

### Networking & Communication

- [API Proxy Gateways](https://awesome-repositories.com/f/networking-communication/api-proxy-gateways.md) — Lets existing agents reach shared memory through a proxy in front of model API requests by changing only the base URL, with no plugin or hook.

### Security & Cryptography

- [Multi-Level Visibility Controls](https://awesome-repositories.com/f/security-cryptography/identity-access-management/access-control/access-control-models/permission-based-security/access-control-policies/project-access-controls/multi-level-visibility-controls.md) — Offers private, team, restricted, and agent-targeted visibility with ownership checks, user, role, and agent access lists, and an admin panel for teams and bindings.

### Part of an Awesome List

- [Agent Memory and State](https://awesome-repositories.com/f/awesome-lists/ai/agent-memory-and-state.md) — Local zero-dependency progressive long-term memory layer.