Back to vendors
Fabi.ai logo

Fabi.ai

Visit site
Data analyst agentindependentVerified 2026-07-01

Fabi.ai is an AI data analyst platform whose Analyst Agent writes SQL and Python inside collaborative Smartbook notebooks, then publishes auto refreshing dashboards and pushes insights to Slack.

Fabi.ai is an AI data analyst platform built for lean data teams and operators who need answers faster than a traditional business intelligence stack can deliver them. Its Analyst Agent takes a question in plain English, writes the SQL to pull the right data from a warehouse, and then writes Python for the harder work such as statistical analysis, machine learning models, and custom visualizations. The goal is not to replace analysts but to compress the path from raw data to a shareable insight from hours into minutes, so a small team can support the whole business.

The workspace is built around Smartbooks, which are AI native notebooks positioned as a next generation alternative to Jupyter and Google Colab. Smartbooks add reactive cells, so a change upstream automatically flows through every dependent step, along with real collaboration and built in version control that brings software engineering discipline to analysis. From a Smartbook a user can publish an interactive dashboard or report in a couple of clicks, and those dashboards refresh automatically on a schedule and stay synced with the latest data rather than going stale the day after they are made.

Fabi connects to the sources a growth team actually uses, including data warehouses, databases, Google Sheets, comma separated files, and customer relationship management and go to market applications, and it can merge several sources in memory for a single analysis. Teams keep control through field and table access rules and a semantics layer that the agent can auto learn or that a person can define, and scoped agents can be deployed to work only against approved datasets. Insights do not stay trapped in the tool, since Fabi pushes results out to email, spreadsheets, slides, and Slack, and can fire real time alerts when a metric moves.

Underneath, Fabi lets teams choose their model, running on OpenAI, Anthropic, or a private large language model, which matters for both quality and data control. A plug and play Model Context Protocol server lets people call the analyst from Slack or from whatever chat interface they already use, so insights come to the user instead of forcing a context switch. Fabi.ai offers a robust free tier, with paid plans starting around thirty nine dollars per month per seat and a custom plan for larger teams. It fits organizations that want a fast, code backed, and governable AI analyst rather than a black box chatbot.

Vendor details

Canonical URL

https://fabi.ai

Category

Data analyst agent

Company status

independent

Use cases & customers

In practice

A lean growth team connects its warehouse and customer relationship management data to Fabi, then asks the Analyst Agent plain English questions that return SQL, Python analysis, and a publishable dashboard within minutes.

An operator builds a Smartbook with reactive cells and version control, publishes it as a scheduled dashboard, and lets teammates ask follow up questions through the Fabi agent installed in Slack.

A data lead scopes an analyst agent to only approved tables, connects it through the Model Context Protocol server, and sets real time alerts so stakeholders learn when a key metric shifts.

Sources & related URLs

Research notes

Score 8.0 (4F/8P/2N). AI data analyst platform (Analyst Agent = SQL + Python + ML), Smartbook AI notebooks (reactive cells, version control, collaboration), auto refreshing published dashboards. Fulls: Int (warehouses, DBs, Sheets, CSV, CRM/GTM, merges sources in memory), Trig (Slack, MCP, email, slides, scheduled refresh, real time alerts), Model (OpenAI, Anthropic, or private LLM), Ext (plug and play MCP server, add to any LLM/chat, Slack app, version control). Partials: Orch (multi step analysis workflows, scheduled automation, deployable scoped agents, reactive cells; not autonomous multi agent), Know (semantics context, auto learn or manual, business context grounding; not a formal governed metrics layer), HITL (field/table access rules, scoped agents on approved datasets only; no explicit approval workflow), Sec (DPA/GDPR plus field/table access controls; no SOC 2 or SSO verified in results, medium low confidence, revisit), Obs (built in version control, reactive lineage; no explicit agent run trace/eval logs), Mem (persistent semantics/business context, Smartbook state; no first class long term conversational memory), Pack (dashboard templates plus purpose built GTM/marketing agents), Eval (AI assisted SQL debugging plus version control; no formal eval harness). N: Dep (cloud SaaS, private LLM is model side; no self host/VPC/residency documented), Comp (no browser/computer use). Pricing self_serve: robust free tier, paid from $39/mo per seat (vendor pricing page, public_exact), custom plan. NOTE: stale saasworthy (Dec 2024) claims $199 start and 'no API'; both contradicted by current vendor site ($39/mo, plug and play MCP server) and Toosio Apr 2026; used current vendor figures.

Capability coverage

8.0 / 14 capabilities · 57%

Integrations & Tool CallingFabi connects to data warehouses, databases, Google Sheets, comma separated files, and customer relationship management and go to market applications, merges multiple sources in memory, and its agent writes and runs SQL and Python, so full. Full
Workflow OrchestrationFabi runs multi step analysis and automated workflows with reactive cells and scheduled dashboard refreshes, and can deploy scoped agents, but it is not documented as an autonomous multi agent orchestrator, so partial. Partial
Knowledge Grounding & RAGFabi lets teams add semantics and business context that the AI can auto learn or a person can define, grounding answers in the business, though it is not a formal governed metrics layer, so partial. Partial
Human Oversight & GuardrailsFabi gives control through field and table access rules and lets scoped agents work only against approved datasets, a guardrail, but no explicit human approval workflow is documented, so partial. Partial
Security, Identity & GovernanceFabi documents a data processing addendum for privacy compliance and field and table access controls, but no SOC 2 or single sign on could be verified in available sources, so partial. Partial
Observability & AuditabilityFabi provides built in version control and reactive cells that trace how changes flow through an analysis, but explicit agent run tracing and evaluation logs are not documented, so partial. Partial
Memory & State PersistenceFabi persists Smartbook state and business semantics that the agent reuses, but a first class long term conversational memory across sessions is not documented, so partial. Partial
Deployment & Data ResidencyFabi is a cloud platform and offers a private large language model option on the model side, but no self host, virtual private cloud, or data residency deployment is documented, so not documented. Unable to verify
Prebuilt Agents, Templates & PacksFabi ships dashboard templates and purpose built analyst agents for go to market teams, a starter template set rather than a full marketplace of cloneable agents, so partial. Partial
Triggers & Channel CoverageFabi installs in Slack, exposes a Model Context Protocol server for any chat interface, pushes results to email, spreadsheets, and slides, refreshes dashboards on schedule, and fires real time alerts, so full. Full
Model Flexibility & RoutingFabi lets teams run analysis on their preferred engine, choosing OpenAI, Anthropic, or a private large language model, so full. Full
APIs, SDKs & MCP ExtensibilityFabi provides a plug and play Model Context Protocol server that adds the analyst to any chat interface, a Slack app, and built in version control and developer tooling, so full. Full
Testing, Debugging & OptimizationFabi assists with writing and debugging SQL and uses reactive cells and version control to iterate on analysis, but a formal evaluation harness for agent output is not documented, so partial. Partial
Browser & Computer UseFabi generates and runs SQL and Python for analysis but has no browser or computer use capability, so not documented. Unable to verify

Recent platform changes

No recent material changes tracked yet.

Pricing

Free tier; from $39/mo per seat

Public — exactLow variable cost
Verified 2026-07-01
Data confidence: medium

Contact us

Found a vendor we missed? Have feedback on the index? We'd love to hear from you.

Agentic AI Index

A directory and comparison resource for AI agent platforms, autonomous workflow tools, and enterprise agentic automation products.

© 2026 Agentic AI Index

3801 N Capital of Texas Hwy, Ste E240 · Austin, TX 78746

Researched from public vendor sources. See Methodology.