Coming soon: Evals. Test a context change before you publish it. See how it works

Bonnard

Scale accurate data querying
for agents, fast

See how well your agents query your data, then improve it with tools and infra out of the box.

Talk to us

Everything your data team needs out of the box

  • MCP server
  • Manage context
  • Versions
  • Evals
  • Feedback loop
  • Data visualisation
  • Agent observability
  • Roles & access

Accurate data for every agent is hard to scale.

Your team built an internal data MCP, wrote context in docs, and shared it internally with great feedback. Rolling it out means evals, context loops, access, tenancy and tools that hold up under real questions.

Everything in three steps.

  1. 01

    Connect your data.

    Point Bonnard at your data, or upload a dbt or OSI file. It reads the tables, keys and joins you already have.

  2. DraftPublishv1 v2Live
    02

    Publish your context.

    Define metrics, rules, evals and metadata from the agent that already knows your business. Draft it through the MCP, review it, then publish it.

  3. MCP BonnardSchema discoveryEvalsActivity recordAccess per source
    03

    Bonnard does the rest.

    The context loop, the MCP and everything around it, so your people and your customers can just get started.

Works with Postgres, BigQuery, Redshift, ClickHouse, SQL Server, MySQL, MotherDuck and PostHog.

Try it on demo data

Let us take care of the infrastructure; you handle the context.

Context management, versions, evals, the context loop, usage records, agent tools, charts and dashboards, sign-in, roles and customer separation, in one MCP.

  • Context when the agent needs it.

    explore_schema hands the agent the right tables and definitions as it asks. Then it queries, charts and builds dashboards.

    Context

    Which products brought in the most revenue in August?

    Explore schema
    • revenuePaid orders only.
      Metric
    • ordersOne row per order.
      Table
    • order_itemsOne row per product line in an order.
      Table
    • productsOne row per product in the catalogue.
      Table
  • One MCP on the data models you have.

    Connect a database and Bonnard serves its context to every agent your people use. No new data model.

    MCP and tools
  • Evals before you publish.Coming soon

    Test a context change against the live version, question by question, and see what got better.

  • A loop that keeps it right.

    Bonnard reads how agents used your data, finds what’s wrong and asks your data team. Their answers go into the context.

  • Roll out with roles.

    Give people and customers access per data source, from the same MCP.

    Governance

All your BI needs through three tools

  • sql_query

    Runs one read-only SQL SELECT and returns the rows.

    Show me every order line from August.

    A table returned in the chat: August order lines with order_id, ordered_at, product and units. The preview shows the first 250 of 4,120 rows.
    A table returned in the chat: August order lines with order_id, ordered_at, product and units. The preview shows the first 250 of 4,120 rows.
  • render_chart

    Draws a chart, or a carousel of up to ten, right in the chat.

    Which products brought in the most revenue in August?

    A bar chart in the chat, revenue by product for August: trail shoes first at about 96,800, then running jacket, base layer and cap.
    A bar chart in the chat, revenue by product for August: trail shoes first at about 96,800, then running jacket, base layer and cap.
  • render_dashboard

    Shows a dashboard in the chat as an interactive grid of tiles.

    Show me the weekly sales dashboard for August.

    A dashboard on a phone, tiles in one column: units by product, then revenue by week.
    A dashboard on a phone, tiles in one column: units by product, then revenue by week.
    A dashboard in the chat with four tiles: units by product, revenue by week, orders and top customers.
    A dashboard in the chat with four tiles: units by product, revenue by week, orders and top customers.

More tools

  • save_chartSaves a chart, or edits one already saved.
  • save_dashboardSaves a dashboard, or edits one already saved.
  • create_export_fileCreates a CSV or PNG without rerunning the query.
  • explore_chartsLists the charts a person saved.
  • explore_dashboardsLists the available dashboards.
  • explore_docsHow-to docs for charts, dashboard params and KPIs.

More chart types

  • Area chart, drawn by the product
    Area chart, drawn by the product
    Area
  • Pie chart, drawn by the product
    Pie chart, drawn by the product
    Pie
  • Scatter chart, drawn by the product
    Scatter chart, drawn by the product
    Scatter
  • Funnel chart, drawn by the product
    Funnel chart, drawn by the product
    Funnel
  • Waterfall chart, drawn by the product
    Waterfall chart, drawn by the product
    Waterfall
  • Bar and line chart, drawn by the product
    Bar and line chart, drawn by the product
    Bar and line

Secure by default, ready for your customers

Read-only from the first query.

  • Sign-in

    OAuth 2.1, PKCE mandatory.

    Tokens are audience-bound per server.

    Bring your own IdP.

    OAuth, API keys or trusted JWTs. SSO supported.

  • Data

    AES-256-GCM at rest.

    Credentials encrypted. API keys stored only as hashes.

    Read-only.

    SELECT statements only. On SQL Server, the login’s grants enforce it. Connects over SSL from fixed IPs.

  • Tenancy

    Set per database.

    Row-level security, or a database or dataset per tenant.

    Tenant from the sign-in.

    An IdP sign-in maps to the right tenant. Scale and Enterprise plans.

No per-user fees, everyone’s included.

You pay per call, however many people make them.

  • Free

    $0

    3,000 calls a month

  • Team

    $150

    20,000 calls a month

  • Scale

    $600

    100,000 calls a month

  • Enterprise

    Custom

    Custom volume

Start scaling
accurate data today

Free for your first 3,000 calls. No seat limits.

Talk to us