MinusX

MinusX

MinusX is an open-source agentic business intelligence platform built from the ground up for AI agents rather than bolted onto a legacy BI tool. It connects to Postgres, BigQuery, Athena, DuckDB, Google Sheets and CSV uploads, then handles the full analyst workflow: ad-hoc questions, SQL analysis, dashboards, metrics, alerts, slides and data stories. Its differentiator is a self-improving knowledge base plus Evals. The agent auto-learns company-specific definitions (what ARR means, which test orders to exclude, fiscal year rules) from every interaction, and you can test answers before trusting them. MinusX claims #1 SOTA on UC Berkeley's DataAgentBench. Also ships Slack Q&A, an MCP server for Claude/ChatGPT-style agents, proactive alerts and scheduled reports. Self-host the Apache 2.0 core with your own LLM keys, or use the managed cloud. Aimed at founders, data teams and non-technical operators who need answers without writing SQL.

Analytics
Visit website
Added on
Nov 4, 2026
Agentic BIOpen sourceApache 2.0Self-hostBYOKSlack botMCP serverData storiesProactive alertsEvals
Product information

Everything worth knowing about MinusX

The complete picture — from everyday features to the technical detail builders and IT folks dig for.

Core features

What it actually does

Agentic BI, not a chatbot
Ask ad-hoc questions in plain English and get real analysis back — SQL, charts, dashboards, metrics, alerts and docs. The agent covers the ~75% of analyst work that isn't writing SQL.
Slides & Data Stories
Turn any analysis into a boardroom-ready deck or a scrollable scrollytelling report — charts, annotations and takeaways drafted end-to-end by the agent.
Proactive alerts, reports & nudges
Scheduled KPI reports, threshold alerts and proactive nudges. MinusX watches your data and pings you in Slack when churn spikes or spend blows past forecast.
Knowledge Base & Evals
The agent auto-learns your business context — what ARR means, which test orders to exclude, your fiscal year — and you can test it with Evals so you know the answers are trustworthy.
Slack, MCPs & Skills
Ask questions from Slack, or expose the MCP server so Claude, ChatGPT and your other AI agents can query your warehouse through MinusX.
Technical capabilities
Self-host or cloud, BYO LLM keys
Apache 2.0 core you can inspect and run in your own environment, or a managed cloud runtime. Bring your own OpenAI/Anthropic/Google keys for 50% off any plan — you control model costs.
SQL GUI, viz builder & query caching
Full SQL editor, point-and-click query builder, advanced filters, maps/funnels/heatmaps, plus advanced query caching for team-scale performance.

Who uses it

Real use cases, no hype
Ad-hoc Data Q&A
Ask natural-language questions and get SQL, charts, and answers in app or Slack.
Auto Dashboards
Build and refine dashboards from warehouses, Sheets, or CSVs with a SQL GUI.
Proactive Alerts & Reports
Schedule KPI reports and trigger threshold alerts when metrics break.
Slides & Data Stories
Turn analyses into board-ready decks and scrollable narrative reports.
Business Context Learning
Teach metric definitions, docs, and exceptions via a shared knowledge base.
MCP Agent Integration
Expose MinusX analytics to Claude, ChatGPT, and other AI agents via MCP.
Cross-functional BI
Run growth, finance, product, ops, marketing, and B2B dashboards in one place.
What's great
  • Open-source core, self-hostable
  • Learns company-specific context over time
  • #1 on DataAgentBench (vendor claim)
  • Slack and MCP integrations built in
Technical strengths
  • Evals framework tests agent answers
  • Knowledge base stores metrics, docs, and corrections
  • Full SQL GUI plus query builder
Where it falls short
  • Young product; connector list is still short
  • Cloud agent credits are hard to budget
  • Self-hosting means you own infra and upgrades
  • SSO/SAML only on Enterprise
Technical limitations
  • Answer quality depends on curated metrics/docs
  • Token-based credit model adds cost opacity
  • Evals/debugging tools are Team-tier, not OSS
For technical folks

The deep-dive specs

License
Apache 2.0
Deployment
Self-hosted (private runtime) or managed MinusX Cloud; dedicated/on-prem for Enterprise
Benchmark
#1 on DataAgentBench (UC Berkeley EPIC lab, as of Jun 2026)
Compliance
SOC2 Type II, GDPR
LLM providers
OpenAI, Anthropic, Google and others; BYOK supported
Data sources
Postgres, BigQuery, Athena, DuckDB, Google Sheets, CSV uploads, Supabase

FAQ

Includes technical Q&A
MinusX is open source (Apache 2.0), so you can self-host and audit the code. The company says it never stores raw data or uses it for ML training; cloud users should still review the privacy policy.

Traffic Insights

Monthly Visits

—

November 2026

Growth Rate

0%

vs last month

Dominance

0%

in Analytics

Top Country

—

Top Countries Breakdown

—

Growth Trend

Stable

Traffic has remained stable.

Honest pricing

No sneaky tiers, no “contact sales”

Here's who each plan is actually for — and where the hidden charges might hit.

Open Source
Free

Developers & small teams with their own infrastructure

  • Self-hosted deployment; Bring your own LLM keys; Core BI features; Open context manager; SOTA data agent
Founders
$40 / user/mo

Solo founders & small teams

  • Cloud-hosted & managed; 500 agent credits / user / month; Proactive analytics & alerts; Self improving context; Multiple SOTA agents
Team
Most picked
$600/mo

Growing teams that need more scale and collaboration

  • Everything in Founders; 10k pooled agent credits; Team workspaces & sharing; Evals and agent debugging tools; Advanced query caching
Enterprise
Custom

Enterprise (typically > 20 users)

  • Everything in Team; Forward Deployed Engineers; Custom data integrations; Dedicated / On-prem deployment; SSO / SAML
View Full Plans & Pricing on MinusX

Official pricing, checkout, and plan upgrades are handled directly on minusx.ai.

Reviews

What the internet actually thinks

Data refreshed weekly. No paid placements.

Trustpilot
Not reviewed yet
G2
G2
Not reviewed yet
C
Capterra
Not reviewed yet
Product Hunt
Not reviewed yet
COMMUNITY COMMENTS

Leave a Comment

Help others make informed decisions. Your honest feedback shapes the community's understanding of this tool.

Tips for commenting

  • •Be specific about features you used
  • •Share real use cases and results
  • •Mention both pros and cons
  • •Keep it honest and constructive
to leave a comment

Your comment is published immediately.

Filter:
Sort:

No comments yet

Be the first to share your experience with this tool.

Alternatives

Not quite the right fit?

Here's what else is out there.