Budibase
Also known as: Budibase, budibase.com, Budibase Agents, Budibase AI, Budibase Automations
Open source operations platform with an agent builder: agents use granted workspace tools, run as the requester or an admin, deploy to Slack, Teams and Discord, and route risky actions to approval policies with any, majority or unanimous sign off.
Budibase is an open source operations platform that puts an agent builder beside its app builder, automation builder and data tables, built for operations and IT teams handling requests, approvals and repetitive workflows. Every workspace has an Agents tab where a builder writes natural language instructions, attaches a model, and grants the agent specific tools. Those tools can be table operations such as list, get and update row, REST API requests, integrated apps, knowledge sources and automations. An agent can only use what it has been granted, and each tool runs either with the permissions of the person asking or with elevated admin rights.
Any tool can carry an approval rule, either on every call or only when its inputs meet a condition, such as a new row whose total exceeds a threshold. Rules point to reusable policies that name the approvers, whether any one, a majority or all of them must agree, and the Slack or Microsoft Teams channel where the approval card appears with the parameters the approver needs. Agents triggered from automations pause on escalation and resume after approval.
Agents reach work through messaging channels in Slack, Microsoft Teams and Discord, and through automations that start on schedules, webhooks and data events and can call an agent as a step. Because automations can also be handed to agents as tools, multi agent flows can be built with deterministic steps around them. Knowledge retrieval runs on a pgvector database with a chosen embedding model, tunable similarity and chunk count, and uploaded files.
Budibase runs on its own cloud or self hosted on Docker, Kubernetes or on premises, with an air gapped option for enterprise customers, and the open source edition includes unlimited agents. The agent builder is labeled beta. The testing guide sets out a regression method that the builder runs in a preview chat, with role simulation and mock approvals, and nothing stores or scores the test cases.
Vendor details
Canonical URL
https://budibase.com
Category
Agent builder
Subcategory
Agent builder inside an open source operations platform
Funding status
Private (Budibase, Inc.), with its open source core published on GitHub. Funding rounds are not stated on its own pages.
Company status
independent
Use cases & customers
Primary use cases
Target customers
Deployment options
Integrations
Agents use as tools any table, REST API request, integrated app, knowledge source or automation in the workspace, with Budibase DB, SQL databases, a custom REST connector with configurable auth, and datasource plugins. Agents deploy to Slack, Microsoft Teams and Discord. Model providers are connected with the customer's own keys, and knowledge retrieval runs on pgvector with a choice of embedding model. A public REST API covers apps, users, tables and data.
In practice
An IT team deploys a service desk agent to Slack that reads and updates ticket rows, with any update to a priority ticket routed to two approvers before it commits.
An operations group calls an agent from a scheduled automation to classify new requests and hands escalations to a majority approval policy in Microsoft Teams.
A regulated company runs the open source edition on its own Kubernetes cluster with pgvector knowledge retrieval over internal policies.
Sources & related URLs
Related / legacy domains
Research sources
Agentic Index coverage score
11.5 / 14 capabilities · 82%
| Integrations & Tool Calling | Full |
|---|---|
|
Any table, REST API request, integrated app, knowledge source or automation in the workspace can be granted to an agent as a tool, Budibase discovers operations from connected data sources, and the custom REST connector carries configurable auth. A worked example lists, gets and updates ticket rows. Sourcedocs.budibase.com/docs/agent-building-101read 2026-10-02 |
|
| Workflow Orchestration | Full |
|
The automation builder chains triggers and action steps, agents are callable as steps inside automations, and automations can be handed to agents as tools, which is how Budibase builds multi agent workflows with deterministic control. Synchronous automations return results to callers. Sourcebudibase.com/blog/updates/ai-agents-betaread 2026-10-02 |
|
| Knowledge Grounding & RAG | Full |
|
Knowledge bases are configured on a pgvector database, assigned to an agent operation with a chosen embedding model, tuned by minimum similarity and chunks to retrieve, and filled with uploaded files and knowledge content that persist for semantic retrieval. pgvector is the only supported provider. Sourcedocs.budibase.com/docs/agent-vector-databases.mdread 2026-10-02 |
|
| Human Oversight & Guardrails | Full |
|
Each tool can carry an unconditional or conditional approval rule, for example one that applies only when an input field exceeds a threshold, so the action waits for approval before it commits. The rule is tied to a reusable policy naming the approvers and whether any one, a majority or all must agree, and approval cards in Slack or Microsoft Teams show the selected parameters and the requester. Agents triggered by automations pause and resume after approval. Tools also run as the requester or an elevated admin. Sourcedocs.budibase.com/docs/agent-tools-and-permissions.mdread 2026-10-02 |
|
| Security, Identity & Governance | Full |
|
Access runs on role based access control with tenant and app roles, user groups, SSO with PKCE OIDC and enforceable SSO, SCIM directory sync on Enterprise, and agent tools that run under the requester's permissions with metadata and results redacted when the requester lacks access. Budibase describes itself as ISO 27001 certified and GDPR compliant and has a security page linked from its site navigation. Audit logs are available on Enterprise. Sourcebudibase.com/pricingread 2026-10-02 |
|
| Observability & Auditability | Full |
|
An Activity timeline records agent tool calls and detailed failure information, agent logs are retained from 1 day on Pro to unlimited on Enterprise and the open source edition, automation runs are inspectable in a logs view, and approval cards name the requester and tool. Platform audit logs sit on Enterprise. The depth of reasoning traces is not described. Sourcedocs.budibase.com/docs/agent-tools-and-permissions.mdread 2026-10-02 |
|
| Memory & State Persistence | Partial |
|
Conversation state lasts within a session. Slack threads act as conversation boundaries with an optional idle timeout, and preview chat history persists in browser session storage for testing. Budibase does not describe a memory layer with a set scope and lifetime across conversations, and the workspace tables the agent reads are the application's data model. Sourcedocs.budibase.com/docs/messaging-channels-slackread 2026-10-02 |
|
| Deployment & Data Residency | Full |
|
The open source edition and Enterprise self host on the customer's own infrastructure or on premises, Business adds a self host on premise option, and air gapped deployment with an offline license is an Enterprise add on; Budibase Cloud is the managed alternative. Sourcebudibase.com/pricingread 2026-10-02 |
|
| Prebuilt Agents, Templates & Packs | Partial |
|
Customers assemble their own agents, with a structured instruction template in the agent builder and a step by step guide to building a service desk agent. No catalog of prebuilt, installable agents is published. Sourcedocs.budibase.com/docs/agent-building-101read 2026-10-02 |
|
| Triggers & Channel Coverage | Full |
|
Agents run autonomously through automations, which start from 20+ built in triggers including webhooks and data events and can call an agent as a step. Agents triggered this way run and pause on escalation without a person starting them. Agents are also deployed to Slack, Microsoft Teams and Discord through messaging channels, with Slack threads as conversation boundaries. Sourcebudibase.com/pricingread 2026-10-02 |
|
| Model Flexibility & Routing | Full |
|
Agents Config connects the customer's own model providers and credentials (bring your own AI key on every plan), sets a default model, and attaches a model per agent. Embedding models for knowledge retrieval are chosen separately. Budibase AI credits are an alternative on Cloud. Sourcedocs.budibase.com/docs/airead 2026-10-02 |
|
| APIs, SDKs & MCP Extensibility | Full |
|
The public REST API works with apps, users, tables and data, with an expanded API on higher tiers, endpoints published in OpenAPI through the docs llms.txt, webhook automations that outside systems call, a CLI, and UI, datasource and automation plugins. Sourcebudibase.com/pricingread 2026-10-02 |
|
| Testing, Debugging & Optimization | Partial |
|
Evals are listed among the plan features, and the testing guide has the builder assemble a prompt set, define pass criteria and rerun it after any prompt or tool change, using a preview chat that simulates roles through Test as and offers mock approve and reject on escalations. Nothing stores, scores or gates the test cases, and the regression comparison is manual. Sourcedocs.budibase.com/docs/agent-testing-guide.mdread 2026-10-02 |
|
| Browser & Computer Use | Not documented |
|
Agent tools are tables, REST requests, integrated apps, knowledge sources and automations, and agents do not control a browser, desktop or remote computer. Sourcedocs.budibase.com/docs/agent-tools-and-permissions.mdread 2026-10-02 |
|
The Agentic Index coverage score grades every vendor Full, Partial or Not documented against the same 14 buyer facing capabilities, from public evidence only. Each capability links to how all vendors in the index score on it. How this evidence is graded
Product updates
Budibase v3.47.0 adds approval policies for agent tools and turns on tool escalations for all users. A policy names the approvers, the voting rule (any approver, unanimous or majority) and the Slack or Microsoft Teams channel, and can apply to every call of a tool or only when the input meets a condition, such as a new row with a total above 1,000.
Bears on: Human approval / guardrails
View sourcePricing
Pro is $19 a month billed yearly, or $22.80 billed monthly. The open source edition is free to self host.
Monthly actions (agent tool calls, automation steps and row writes), plus creator and end user add ons. Cloud tiers also carry AI credits.
Included quota
Pro includes 5K or 10K actions depending on the billing view, 2K Budibase AI credits, 1 creator, 1 day of logs, 1 workspace and unlimited agents. Premium includes 20K or 50K actions, 10K AI credits and SSO. Business includes 250K actions, 50K AI credits, 3 creators and 30 days of logs.
What is public
All four Cloud tiers, the free open source and Enterprise self hosted editions, add on prices, action and AI credit allowances, log retention and a full feature comparison are published. Enterprise is custom.
Billing mechanics
Each Cloud tier is a flat monthly price set by actions, billed yearly or monthly, with monthly billing 20 percent above yearly. Creator and end user add ons follow the same billing period. Cloud tiers are paid by card through Stripe, and Enterprise is invoiced annually.
Cost watchouts
End users are an add on on every Cloud plan ($5 each a month, none included), so a team with many app users pays well above the headline. Audit logs, SCIM and air gapped deployment are Enterprise only. Actions are hard limits, so a busy agent stops at the cap until the plan is raised.
Variable cost rationale
Actions are the meter. Each agent tool call, automation step or row write counts as an action, and every Cloud tier carries a hard monthly allowance, so growth means moving up a tier rather than paying overage. End users at $5 each a month and extra creators at $50 grow the bill with the audience and the build team. The limits are firm. Tier prices are published, actions stop at the cap with no overage billing, and bringing your own model key moves model cost to the customer's provider. The open source edition self hosts free with unlimited agents and users.
Additional watchouts
No end users are included on Cloud tiers, so the headline price understates the cost of an app with a large audience.
Overage / add-ons
No overage is billed. Actions are hard monthly limits, and when they run out, agents and automations stop until the workspace moves to a higher tier.
Sales call required
Mixed (some tiers require a call)
Free / trial
Free open source self hosted edition with unlimited agents, and a 14 day free trial on Cloud plans
Lowest paid plan
Pro at $19 a month billed yearly, or $22.80 billed monthly
Commercial notes
Cloud tiers have a self serve trial and checkout, and Enterprise is sold through sales. The open source edition is free indefinitely.
Key ambiguities
Pro and Premium each show two action figures (5K and 10K, 20K and 50K), and neither is labeled as the yearly or the monthly allowance.
Cancellation / refund
Cloud tiers bill yearly or monthly by card through Stripe, and Enterprise is invoiced annually. No cancellation or refund terms are published.
Support SLA / resale
Community support on the open source edition, Pro and Premium, and email support on Business. Enterprise adds priority support with a 2 business day SLA.
Missing data
Which action figure applies to which billing period on Pro and Premium, and the Enterprise floor.
Related vendors
- Lindy — Lindy is an AI teammate that lives in Slack, answers from your tools…
- Activepieces — Open source, MIT licensed automation platform where every one of 700…
- Agenta — Agenta is an open source workspace where teams build AI agents by…
- AgentX — No code platform to build, evaluate, and deploy multi agent…
- AI Library — No code platform for building and running AI agents, with a…
- Aigensei — Enterprise agentic workflow platform that runs business processes…
Alternatives to Budibase
The closest documented capability profiles to Budibase among agent builders tracked by Agentic Index, ordered by similarity on the same 14 point evidence the rankings use. No vendor pays for placement.
- Activepieces11.5 / 14Matches Budibase across all 14 documented capabilities
- Cohere North11.0 / 14A lighter documented profile than Budibase
- Dify12.0 / 14Fuller documented coverage on Prebuilt Agents, Templates & Packs
- Joget11.0 / 14A lighter documented profile than Budibase
- LuMay12.0 / 14Fuller documented coverage on Prebuilt Agents, Templates & Packs
- Retool12.0 / 14Fuller documented coverage on Testing, Debugging & Optimization
Similarity is computed from each vendor's Agentic Index coverage score evidence, axis by axis, not from the totals. How this evidence is graded