Connect, govern and serve your business knowledge
From data connectors to permission-aware MCP access, AgentBrain is the data and context layer that people and AI assistants share.
Data Connectors
Connect the systems where business data already lives and query or extract from them under the same permission model.
- Databases: PostgreSQL, MySQL, BigQuery, MongoDB, Cloudflare D1, SQL Server (query only)
- Files and storage: Google Sheets, Google Drive, Amazon S3, Cloudflare R2, Google Cloud Storage
- HTTP endpoints and Redis, with encrypted stored credentials
Knowledge Base & Search
Store documents and media in folders with permissions, then find them with full-text search refined by semantic ranking.
- PostgreSQL full-text search with vector re-ranking
- Embedding provider of your choice, including self-hosted OpenAI-compatible endpoints
- Folder and item permissions applied to every search result
Knowledge Graph & Provenance
Model entities and relationships across your knowledge, with the evidence behind each fact kept next to it.
- Validity windows (valid from / valid until) on entities
- Extraction provenance and audit history per entity
- Relationship modelling across documents and data sources
Access Governance & Audit
One permission model covers folders, knowledge, media, connectors, graph entities and workflows, and important actions land in an append-only audit log.
- Grants to users, roles and groups, scoped by organization and department subtree
- Row filters on data access and approval rights on sensitive steps
- Append-only audit log enforced in the database, with configurable retention
ETL & Data Workflows
Move and reshape business data with scheduled or webhook-triggered workflows: extract, transform, load, and feed results back into the knowledge base or graph.
- Extract, filter, aggregate, transform and load nodes, plus sandboxed JavaScript transforms
- Knowledge base and knowledge graph input and output nodes
- Human approval checkpoints before sensitive steps continue
MCP & API Access
AI assistants reach governed context through an MCP server that asks the backend to check permissions before returning anything.
- MCP tools for search, context retrieval, reading and managing resources
- Permission checks run in the AgentBrain backend, not in the assistant
- Raw SQL execution only for organizations that enable it explicitly
Agents and channels connect around AgentBrain
AgentBrain is the governed context layer. Agent runtimes and chat channels are integrations, and each one below is labelled with what exists today.
- MCP clients (Claude, Cursor and other assistants)Implemented
Connect any MCP-capable assistant to the AgentBrain MCP server for permission-checked search and retrieval.
- DeweeSeparate product
Separate agent execution product. Connects through MCP with per-deployment configuration.
- GoClawSeparate product
Agent runtime from the same team. AgentBrain does not depend on it; it can connect as an MCP client.
- Agent tasks inside data workflowsImplemented
Workflows can include agent task and LLM call steps. Agent tasks over MCP are not supported yet.
- Telegram and Lark notificationsImplemented
Workflows can send outbound notifications. AgentBrain does not run inbound chat bots.
- Slack, Microsoft Teams, LINE, ZaloNot supported
No native channel integration in AgentBrain today.
Ready to give your business knowledge one governed home?
Book a demo or talk to our team about your data sources and access rules.