AgentSpot

Announced on October 22, 2026, AgentSpot is ThoughtSpot’s pitch for what comes after the chatbot era: “Multiplayer AI.” The idea is simple — instead of one person chatting with one AI assistant, anyone in a company can use plain language to build AI agents, recurring workflows and live interactive apps, then share them in a shared workspace so that “what one person builds becomes a trusted, reusable capability for the entire enterprise.”

Key facts: three building blocks — Agents (AI teammates that investigate, analyze, recommend and act within set permissions), Workflows (recurring tasks on a schedule or trigger, like weekly sales forecasts or onboarding sequences), and Apps (live interactive apps built from a prompt, combining company data with agent actions). Under the hood is a four-layer memory model (company, agent, user, agent-user) and ThoughtSpot’s agentic semantic layer — shared business definitions like “revenue” or “at-risk account” that keep every agent speaking the same language. Enterprise governance is baked in: permissions so users only ever see data they’re authorized for, always-on audit trails, and connectors to Salesforce, Slack and more (the launch demo had an agent pull Salesforce pipeline data and post a summary to Slack).

Plans: there is a genuine free plan — no time limit, no credit card required, with unlimited agents and unlimited users on every tier. The free tier runs on a weekly allowance of agent conversations, workflow runs and app sessions; building and testing is free. The paid “Fleet” tier lifts the caps, but per-unit prices are not published — the model is pay-per-outcome (per conversation, workflow run or app session) rather than per token or per seat.

Best for: teams and companies that want shared, governed AI agents grounded in their own data — sales, finance, marketing, HR, legal and ops. Watch out: the free tier is capped by weekly usage allowance, and with per-unit prices unpublished, heavier users will need a sales conversation before they can predict costs. ThoughtSpot is a Gartner Magic Quadrant leader with 1,000+ customers including Lyft, Unilever, Roche and HubSpot, so this isn’t a startup gamble — it’s the enterprise data platform moving into the agent game.

Related: Gamma · Notion AI · Kore.ai AutoLoop

Pros

• Anyone can build AI agents, recurring workflows and live apps in plain language, no code required.
• Built work is shared across the company, so one person's agent becomes a reusable capability for the whole team.
• Enterprise governance is baked in, with permissions, always-on audit trails and a shared semantic layer for business definitions.
• Connectors cover tools such as Salesforce and Slack, and the free plan needs no credit card.

Watch out for

• The free tier runs on a weekly allowance of agent conversations, workflow runs and app sessions.
• Paid per-unit prices are not published, so heavier users will need a sales conversation before they can predict costs.
• Announced on October 22, 2026, it is a new product, so its ecosystem is younger than established agent platforms.

Best for

Teams and companies that want shared, governed AI agents grounded in their own data, across sales, finance, marketing, HR, legal and ops. Who should skip it: solo users who just want a simple personal chatbot.

Pricing

There is a genuine free plan with no time limit and no credit card required, offering unlimited agents and unlimited users on every tier within a weekly usage allowance. The paid Fleet tier lifts the caps on a pay-per-outcome model, but per-unit prices are not published, so check the official pricing page.

FAQs

What is AgentSpot? ThoughtSpot's platform for building AI agents, workflows and apps in plain language and sharing them company-wide.

What are the three building blocks? Agents, Workflows for recurring tasks, and Apps built from a prompt.

Is there really a free plan? Yes, with no time limit and no credit card required, within a weekly usage allowance.

How is data access controlled? Through permissions so users only see data they are authorised for, plus always-on audit trails.