

TencentDB Agent Memory
β Text source-checked Nov 16, 2026A team-level memory hub for AI agents that turns conversations, docs, and code into four reusable memory assets.
Source-checked against the vendor's site Β· rebuilt from the vendor's own pagesProfile updated Link checked How we rate
TencentDB Agent Memory is a team-level memory hub for AI agents that converts conversations, documents, and code into four reusable memory assets: Chat Memory, Skill, LLM-Wiki, and Code-Graph. The memory assets are decoupled from agent frameworks, allowing them to move across frameworks and be shared and maintained by multiple agents and team members.
Chat Memory retains user preferences, facts, decisions, and interaction history. Skills include versions, resource files, trigger boundaries, execution steps, and validation rules rather than being simple prompt snippets. The Wiki component turns product docs, design specs, and ops runbooks into structured pages with a link graph. CodeGraph indexes code symbols, files, call relationships, and impact paths to support impact analysis before code modification.
Retrieval combines BM25 and vector search with Reciprocal Rank Fusion to fall back to lower memory layers when specific facts are needed. Integration is handled through a proxy that allows agents to connect by pointing their base URL to the service without requiring plugins or hooks. The tool supports integration with DeepSeek Harness, Claude Code, Codex, CodeBuddy, WorkBuddy, Hermes, and OpenClaw.
CodeGraph supports public HTTPS repositories and private repositories with configured credentials.
According to TencentDB Agent Memory: TencentDB Agent Memory is a team-level memory hub for AI agents that converts conversations, Documents, and code into four reusable memory assets: Chat Memory, Skill, LLM-Wiki, and Code-Graph. Its pages list this strength: Memory assets are portable and compatible with multiple agent frameworks.
β The AI Tool Bible editorial team
Pros
- β Memory assets are portable and compatible with multiple agent frameworks.
- β New memory assets are private by default, requiring explicit action to share.
- β The system supports cold-start by importing existing documents, codebases, and conversation sessions.
- β The vendor says that integration requires zero-code changes, only pointing the agent's base URL to the proxy.
Facts
- Platforms
- api (not source-checked)
- API available
- Yes (not source-checked)
- Open source
- Yes (license: NOASSERTION) (not source-checked)
- GitHub stars
- 30,834 (checked Nov 13, 2026) (not source-checked)
- Last GitHub push
- Nov 10, 2026 (not source-checked)
- First commit
- Apr 2026 (not source-checked)
- Company
- TencentCloud (not source-checked)
- Model used
- Multi-model (not source-checked)
- Vendor pages
- Changelog