🧠 A BiSkilled Product · Live now · Commercial

The semantic layer that lets AI agents query your data — safely.

AgentData connects to your existing databases read-only, auto-discovers your business entities, and builds one multilingual semantic model. Humans and AI agents then ask questions in plain language over REST + MCP — no SQL, no data movement.

Only the model and the question ever reach the LLM — never your row data. Self-hostable and air-gap ready.

Book a Demo Open the Live App ↗ Connect via MCP →

In one sentence

AgentData is a self-hostable semantic layer and data fabric that auto-discovers the business entities inside your databases and serves them to people and AI agents as plain-language questions over REST and MCP — without writing SQL and without your row data ever leaving your environment. It's a commercial data-management product, with integration pipelines on the roadmap.

Why teams reach for AgentData

The gap between your data and the AI agents that want to use it

🔌

"Our data is in five different systems"

Postgres here, Snowflake there, an S3 lake, a legacy SQL Server. AgentData clusters equivalent tables across all of them into one business entity.

🧩

"Modelling our semantic layer by hand takes months"

AgentData profiles every table, classifies it into an entity and role, infers relationships, and emits Cube + dbt-semantic YAML you simply review and approve.

🔐

"We can't send row data to an LLM"

Only the semantic model and the question reach the LLM. SQL runs locally against read-only sources; only the result returns. Run it fully air-gapped if you need to.

🤖

"We want agents to answer data questions"

One validated query path backs both a REST API and an MCP server — so Claude, ChatGPT or your own agents query governed data, not raw tables.

How AgentData works

Two paths: one writes the model, one reads the data — both governed

DISCOVERY — writes the model

Point AgentData at a source and it runs: profile → classify → cluster → relate → emit. Equivalent tables across sources collapse into a single entity (Order, Customer, Product…), relationships are inferred, and a Model Registry is written for human review.

Approve-by-default review, drag-and-drop merges (union / join / SCD), calculated columns, and metrics taught in plain language ("revenue = price × qty − discount").

RETRIEVAL — reads the data

A question runs: plan → validate → dispatch → execute → result. The planner generates SQL from the approved model only, validates it, and executes it locally against your read-only sources.

Conversational memory ("just the top 2 of those"), multilingual questions (ask in Hebrew or any language), and saved-and-approved queries that guide the planner for everyone.

What you get

🗄️

Heterogeneous, read-only sources

One config-driven splitter routes each source to the right adapter and engine — native, Trino or Athena.

  • PostgreSQL · MySQL · SQL Server · Oracle
  • Snowflake · Redshift · Synapse
  • S3 + Glue/Athena · Azure Blob (lake)
  • Read-only adapters, encrypted credentials
🧠

AI-assisted discovery

Profiles, classifies and clusters tables into business entities and emits standards-based YAML.

  • Cross-source entity clustering
  • Relationship inference
  • Cube + dbt-semantic YAML output
  • Heuristic fallback with no LLM
💬

Conversational & multilingual

Ask in plain language, follow up, and ask in any language — answers show the plan, the model YAML and the SQL.

  • Natural-language querying with memory
  • Any language (e.g. Hebrew, RTL UI)
  • Save & approve good queries
  • Named star-schema models in a Graph
🔗

REST + MCP for agents

The same validated retrieval path backs both interfaces — multi-tenant by per-user API key.

  • REST API for apps and dashboards
  • MCP server for Claude / ChatGPT / agents
  • list · describe · query_metric · query_nl
  • Only confirmed entities are queryable
🔐

On-prem & air-gapped

Keep everything inside your boundary. Pluggable LLM backend, zero external egress when you need it.

  • Local Ollama / vLLM / TGI (no egress)
  • AWS Bedrock over PrivateLink
  • Anthropic cloud option
  • Model + question only — never row data
📊

Built to operate

Production concerns handled out of the box, from cost to audit.

  • Per-call LLM cost tracking
  • Self-healing watchdog
  • Full audit log
  • Optional exact cross-source federation (Cube + Trino)

Connect an AI agent in one line

AgentData ships a live, secure Model Context Protocol endpoint. Create a per-user API key in the Connect tab, then point any MCP client at it.

claude mcp add --transport http agentdata \
  https://agentdata.mdm.biskilled.com/mcp/ \
  --header "Authorization: Bearer <YOUR_API_KEY>"

Exposes list_entities, describe_entity, query_metric and query_nl over the same validated path used by the REST API. query_nl works in any language and never hard-fails.

Open the Live App ↗ Book a Demo

How AgentData compares

A modern alternative to static dashboards and hand-built semantic layers (Cube · dbt · timbr) — AgentData auto-discovers your model and serves it to AI agents, not just BI tools.

Static dashboards / BI Hand-built semantic layer
(Cube · dbt · timbr)
AgentData
Setup A new dashboard per question (days–weeks) Hand-model every cube & metric Auto-discovers entities; you review & approve
Ad-hoc questions Pre-defined only Write a new model / query Ask in plain language, any language — instantly
Built for AI agents No Limited / custom Native, over MCP
Many / mixed sources One tool at a time Model each source separately Collapses equivalent tables into one entity
Your data & the LLM Varies Varies Read-only; only the model + question reach the LLM
Deployment Usually SaaS Varies Self-host · on-prem · air-gap

AgentData emits standard Cube + dbt-semantic YAML, so it complements those ecosystems rather than locking you in.

AgentData FAQ

Does AgentData move or copy my data?

No. It connects read-only. Only the semantic model and your question reach the LLM — never row data. SQL is generated from the model and executed locally; only the result comes back.

Which databases are supported?

PostgreSQL, MySQL, SQL Server, Oracle, Snowflake, Redshift and Synapse via SQLAlchemy, plus S3 + Glue/Athena and Azure Blob via a lake adapter.

Can it run fully on-premises or air-gapped?

Yes. Point it at a local OpenAI-compatible server (Ollama / vLLM / TGI) for zero external egress, or use AWS Bedrock over PrivateLink. Credentials are Fernet-encrypted at rest; adapters are read-only.

How is it different from a traditional semantic layer or data catalog?

Instead of hand-modelling every cube and metric, AgentData auto-discovers entities across heterogeneous sources, collapses equivalent tables into one entity, infers relationships, and emits Cube + dbt-semantic YAML you review — then serves it to both people and AI agents through one validated path.

What is a semantic layer?

A semantic layer is a single, governed model of your business — entities like customers, orders and revenue — that sits over your raw database tables so people and tools can ask questions in business terms instead of SQL. AgentData builds and serves that layer automatically, over REST and MCP.

Is AgentData a timbr, dbt or Cube alternative?

Yes — it covers the same need (one governed semantic model over your sources) but auto-discovers entities instead of requiring you to hand-model every cube and metric, and it's built for AI agents over MCP, not just BI. It emits Cube + dbt-semantic YAML, so it complements those tools rather than locking you in.

See AgentData on your own data

Book a 30-minute demo and we'll connect it to a sample of your stack — or open the live app and explore the Northwind demo yourself.

Book a Free Demo Open the Live App ↗