# Botpress — AI agent platform with a visual studio, a code-first TypeScript ADK and a helpdesk, priced per conversation. > Source: The Agents Index — https://theagentsindex.com/botpress (structured, researched, re-verified) > Facts last verified: 2026-09-02 Botpress is a cloud platform for building and running AI agents, with a visual studio, a code-first TypeScript ADK, and a helpdesk console. Its agents answer, call tools and take actions across web, messaging and email channels. | Fact | Value | | --- | --- | | Website | https://botpress.com | | Price | from $189/month billed monthly, $150/month billed annually | | Free tier | Yes | 100 conversations, 3 seats, 3 AI agents, Studio and Desk, Help Center, resolution SLA, community support. Hard-capped: no top-ups, no overages, no storage add-on (1K table rows, 100MB vector, 100MB file), no whitelabel webchat, no WhatsApp, no audit log, and no credit-grant auto-recharge. ## What it does **Writes code** · Partly — AI generates the code inside Studio Execute Code cards and Actions from a prompt; the artifact is bot logic, never source in your own repository. > The Execute Code card allows you to generate code using AI, so you don't have to write it all from scratch > — https://botpress.com/docs/studio/concepts/cards/execute-code · Docs page body, 'Generate code using AI' section, first sentence **Debugs and fixes** · Partly — Agent(0), the assistant in the ADK Dev Console, edits actions and workflows and debugs build and runtime errors in your own ADK project, not an arbitrary repository or a CI failure. > Agent(0) can: Edit your actions and workflows; Run CLI commands; Debug build and runtime errors; Send test messages to your bot; Inspect traces and read logs > — https://botpress.com/docs/adk-v2/ai-native/agent0 · Docs page body, 'Agent(0) can:' bulleted list **Writes tests** · Partly — Tests are evals: automated conversation tests of agent behaviour, run by the adk CLI. The adk-evals skill writes them, and there are no unit tests over source code. > Automated conversation tests for your agent. > — https://botpress.com/docs/adk-v2/testing/evals · Docs page body, page description line above the first section **Reviews code** · No — The assistant's own capability list covers editing, debugging, testing and log reading; reviewing a diff someone else wrote is absent from it, and no vendor page claims code review. > Agent(0) can: Edit your actions and workflows; Run CLI commands; Debug build and runtime errors; Send test messages to your bot; Inspect traces and read logs > — https://botpress.com/docs/adk-v2/ai-native/agent0 · Docs page body, 'Agent(0) can:' bulleted list — the complete enumeration of what the assistant does **Whole-codebase work** · No — The unit of work is one agent project whose primitives each live in a file, not an arbitrary repository carrying a migration. > Each primitive lives in its own file. The ADK discovers them automatically, no manual registration. > — https://botpress.com/docs/adk-v2/get-started/introduction/ · Introduction to the ADK, How it works section, sentence after the primitives table **Ships it** · Partly — One CLI command ships an ADK agent to a production bot on Botpress Cloud; it does not open a pull request or deploy to infrastructure you own. > Ships code and config to your production bot. > — https://botpress.com/docs/adk-v2/cli/cli-reference · Docs page body, 'Project' command table, adk deploy row **Done-for-you service** · Partly — You normally run the software yourself; a separately quoted Managed Plan has Botpress experts build and maintain the agents, and a partner-matching service exists alongside it. > Custom AI agents for customer support, lead generation, and product recommendation—built and maintained by our experts to maximize ROI. > — https://botpress.com/pricing · 'Built and managed for you.' section beneath the plan cards ## How much it does unattended **Runs autonomously** · Yes — Autonomous Nodes run an unattended reasoning loop that decides what to do next and which tools to call; handing off to a human is a configured exit, not the default. > You can use the Autonomous Node to build an independent AI agent that: Aligns with your brand; Learns your users' needs; Doesn't need human intervention to take actions > — https://botpress.com/docs/studio/concepts/nodes/autonomous-node · Docs page body, bulleted list under 'You can use the Autonomous Node to build an independent AI agent that:' **Multi-agent** · No — Botpress 'Agents' are specialised components over one bot that run in a fixed vendor-defined order; the closed list has no orchestrator, peer delegation or sub-agent spawning. > When your bot has multiple Agents enabled, they execute in the following order: 1. Summary 2. Personality 3. Policy 4. Translator 5. Knowledge 6. Vision 7. Router 8. Analytics 9. HITL Agent > — https://botpress.com/docs/studio/concepts/agents/introduction · Docs page body, 'Order of execution' section, numbered list **Agent permissions** · Partly — A human can approve AI output before it is sent, and an Autonomous Node cannot read or write variables until access is granted; no approval gate exists before a tool call executes. > The Human-in-the-Loop (HITL) integration allows members of your Workspace to: Chat directly with users; Approve AI-generated outputs before they're sent; Handle complex or sensitive scenarios > — https://botpress.com/docs/integrations/integration-guides/hitl/introduction · Docs page body, bulleted list of what Workspace members can do ## Which models it runs on **Claude** · Yes — First-party Anthropic integration maintained by Botpress; you pick Claude models per node, workflow or skill, billed at cost to your workspace AI Spend. > This integration allows your bot to choose from a curated list of Claude models from Anthropic as the LLM of your choice for a node, workflow, or skill in your bot. > — https://botpress.com/integrations/anthropic · Integration page body, 'Anthropic Integration' description paragraph **GPT** · Yes — First-party OpenAI integration maintained by Botpress, covering text generation plus image generation and speech-to-text, billed at cost to workspace AI Spend. > This integration allows your bot to choose from a curated list of models from OpenAI for content generation (LLM), image generation (text-to-image), and audio transcription (speech-to-text). > — https://botpress.com/integrations/openai · Integration page body, 'OpenAI Integration' description paragraph **Open models** · Partly — Open-weight models are reachable only through hosted providers Botpress maintains: Cerebras, Groq, Fireworks AI and Mistral AI. There are no local weights, no Ollama, Together or Hugging Face integration, and no self-hosted inference. > This integration allows your bot to choose from a curated list of models from Cerebras for content generation and chat completions. > — https://botpress.com/integrations/cerebras · Integration page body, 'Cerebras Integration' description paragraph **Model choice** · Yes — You choose the models and can mix them across one workflow, per bot or per Autonomous Node; Botpress then routes between your own fast, best and fallback slots, and picks for you only if you set nothing. > Choose and configure the models powering your agent. Mix and match LLMs across different parts of your workflow — optimize for cost, speed, or capability depending on the task. > — https://botpress.com/features/ai-agent-studio · Page body, 'Everything your support team needs.' feature grid, 'Multi-LLM strategy' tile ## Where you use it **In your editor** · Partly — No first-party editor extension: the ADK is code you author in your own editor, and Botpress ships skills and MCP config for Claude Code, Cursor, VS Code and Codex. Studio itself is a browser IDE. > Give AI coding assistants like Claude Code, Cursor, and Codex deep ADK knowledge. > — https://botpress.com/docs/adk-v2/ai-native/skills · Docs page body, page description line above 'Install' **On the command line** · Yes — The adk CLI scaffolds, runs, deploys and debugs agents, and pricing is identical whether you build through the CLI or the visual builder. > The Agent Development Kit (ADK) is a CLI tool and development framework for building AI agents on Botpress. > — https://botpress.com/docs/adk-v2/get-started/introduction · Docs page body, first paragraph under the page title **In your pipeline** · Partly — A bot can subscribe to pull-request and issue events and post comments through the GitHub integration, but there is no GitHub Action or CI runner: shipping is the local adk deploy command. > The app has the necessary permissions to listen to pull request and issue events and to create comments on issues, pull requests, and discussions. > — https://botpress.com/docs/integrations/integration-guides/github · Docs page body, 'Basic configuration' tab, Note callout **In a browser** · Yes — Studio is a hosted browser IDE on Botpress Cloud, alongside the Desk support console and the Viber builder; the ADK also serves a local browser Dev Console. > Botpress Studio is an integrated development environment (IDE) for building, testing, and managing AI agents and chatbots. > — https://botpress.com/docs/studio/introduction · Docs page body, opening sentence ## Whose machine it runs on **Self-hosted** · No — Cloud-only: the self-hostable v12 server is sunset and Botpress Cloud is the go-forward platform. The single exception is that you may self-host your own custom integration. > No. Botpress v12 and all self-hosted versions are no longer available for download, purchase, or new deployments. > — https://botpress.com/docs/studio/guides/advanced/v12 · Docs page body, FAQ answer to 'Can I still download Botpress v12 or a self-hosted version?' **Open source** · Partly — The MIT-licensed monorepo holds the public integrations, the CLI, SDK and API client and example bots; Botpress Cloud, Studio and Desk are not in it. > All packages in this repository are open-source software and licensed under the MIT License. > — https://github.com/botpress/botpress · README, 'Licensing' section, final paragraph ## What it costs to run **How it meters** · Yes — Metered per conversation, not per seat or per token: extra conversations come in packs of 100 and LLM spend is included at $0.10 per conversation, above which credit grants are auto-purchased at cost. > Any exchange with at least two messages from the end user in the billing month counts as a conversation. A conversation that spans two months counts in both months. > — https://botpress.com/pricing · FAQs accordion, 'What counts as a conversation?' answer **Free tier** · Yes — A Free plan gives 100 conversations, 3 seats and 3 AI agents with no top-ups; the homepage states no credit card is required, so no sales call is needed to start. > 100 conversations No top-ups or overages > — https://botpress.com/pricing · Free plan card, allowance line beneath the $0 price **API access** · Yes — Public Chat, Admin, Files, Tables and Runtime APIs at api.botpress.cloud, authenticated with a bearer token; no row in the plan comparison table gates API access. > As a general rule, if a feature is available in the Botpress Dashboard, it's likely accessible via the Admin API. > — https://botpress.com/docs/api-reference · Docs page body, 'Admin API' section, closing sentence **MCP server** · Yes — Helpdesk operations are exposed as MCP tools for Claude Code, Claude Desktop and Cursor, and in the other direction an agent can call tools from an MCP server you configure. > Helpdesk operations exposed as MCP tools — connect internal systems without custom integration work. Works with Claude Code, Claude Desktop, and Cursor. > — https://botpress.com/enterprise · Page body, security section, 'MCP server' card body text **Bring your own key** · Partly — Bring-your-own-LLM is Enterprise-only and sales-gated, and the mechanism is building an SDK integration rather than pasting a key; on other plans Botpress makes and bills the model calls. No bring-your-own-cloud. > Using your own LLM is available on Enterprise plans. To get started with your own LLM, contact our sales team. > — https://botpress.com/docs/integrations/sdk/integration/use-your-own-llm · Docs page body, Note callout above the first paragraph ## Buying it for a team **A company can buy it** · Yes — Workspaces carry billing, invoice history and a Tax ID field; a paid Team plan and a sales-quoted Enterprise tier are both organisational purchases scoped to conversation volume. > Enterprise agreements are scoped to the deployment, not the headcount. Pricing is based on conversation volume — not on how many human agents are on your team. > — https://botpress.com/enterprise · Page body, 'No seat-based pricing. No surprise overages.' section, body paragraph **Seat model** · Partly — Seats are never billed per head but they are capped: 3 on Free, 3 on Plus, unlimited on Team and Enterprise. No minimum seat floor is stated anywhere. > Seats are the number of people who can access and work in your workspace. > — https://botpress.com/pricing · Free plan card, 'Includes:' list, seats line and its tooltip text; per-plan values in the comparison table 'Workspace > Seats' row **Pooled budget** · Yes — Conversation quota, AI spend and storage are metered per workspace rather than per seat, and add-ons apply only to the workspace that bought them. > When your workspace reaches 95% of its AI spend limit, we automatically purchase a $10 credit grant of AI spend on your behalf, sold at cost. > — https://botpress.com/pricing · FAQs accordion, 'How do AI spend credit grants work?' answer, first sentence **Admin controls** · Yes — The Policy Agent gates every interaction against business rules, and inputs can be sandboxed with topics blocked outright; role-based access control, per-agent access control and trusted-domain restriction need Team or higher. > The Policy Agent also introduces control mechanisms that act as gates for bot behaviour. These gates serve as checkpoints, ensuring that every interaction complies with designated business rules > — https://botpress.com/docs/studio/concepts/agents/policy-agent · Docs page body, 'Additional Information' section, third paragraph **Audit log** · Partly — A workspace activity log sits under Settings > Audits and a Manager role may view audit records, but the pricing row for it is empty on Free, since it starts at the Plus plan. > You can view a list of all activity that's occurred in your workspace in Settings > Audits. > — https://botpress.com/docs/studio/get-started/configure-your-workspace · Docs page body, 'View workspace activity' section; per-plan values in the pricing table row 'Access a detailed log of changes made within a Workspace' **Single sign-on** · No — Neither SAML nor OIDC is named anywhere: the Enterprise card's complete list of additions has no SSO line, the pricing table's Security & Compliance rows enumerate five items without it, and the docs sitemap has no SSO page. > Includes everything in Team, plus: Dedicated support Uptime SLA Security review Custom storage Voice channel > — https://botpress.com/pricing · Enterprise plan card, complete 'Includes everything in Team, plus:' list ## What happens to your code **Opt out of training** · Yes — No opt-out is needed: the vendor states outright that it does not use conversation data or uploaded content for model training, and says the same of the provider side. Aggregated usage data is still used for product improvement. > We do not use Conversation Data or other End-User personal data for any other purpose, which includes analytics and algorithms and model improvements or training. > — https://botpress.com/legal/privacy-statement · Section 1 'What Personal Data is Collected Through The Service', End-Users subsection, closing sentence **Data residency** · Partly — Default hosting is AWS in the United States; European Data under the DPA is processed within Europe, but defining custom retention and residency policies is an Enterprise-only row and no region picker is documented. > is stored electronically by our service providers on servers located in the United States or other locations determined by the relevant Customer > — https://botpress.com/legal/privacy-statement · Section 4 'How We Store and Transfer Personal Data', third paragraph **Getting out** · Yes — Conversations and analytics export via the API and the DPA requires deletion or return at termination; unused conversation packs expire at month end while AI credit grants carry for up to 12 months. > Packs persist month-to-month until you remove them in billing settings. Unused conversations in a pack expire at month end — they don't roll over. > — https://botpress.com/pricing · FAQs accordion, 'What happens when I hit my conversation limit?' answer, closing sentences **Certifications** · Partly — SOC 2 and GDPR are claimed with KPMG penetration testing and US/EU/Swiss Data Privacy Framework certification; a DPA comes with Team and up, a HIPAA BAA only with Enterprise, and ISO 27001 is claimed nowhere. > SOC II certified, GDPR compliant, and penetration tested by KPMG. Your data and your customers' data are protected to the highest industry standard. > — https://botpress.com/ · Homepage body, 'Enterprise-grade reliability, built over 10 years.' section, 'Robust security' card body text ## Platform record (as of 2026-08-30) Reported by the code-hosting platform, not by the vendor. | Fact | Value | | --- | --- | | Licence | MIT | | Stars | 14882 | | Forks | 2300 | | Last push | 2026-08-28 | ## Provenance 1 of 46 sources behind the facts above are independent of the vendor. ## Coverage elsewhere Pages about this product that the vendor does not control. The label states how independent each one is. | Source | Kind | Read | | --- | --- | --- | | betakit.com | Independent source | 2026-08-23 | | productgrowth.in | Independent source | 2026-08-23 | ## What changed - 2026-09-02 — Capabilities · Commercial terms · Free tier · Price - 2026-08-31 — Capabilities · Commercial terms · Free tier · Price - 2026-08-23 — Alternatives · Documentation link · One-liner · Summary - 2026-08-19 — One-liner · Summary ## Correcting this record Anything marked "Not established" is a question we have not answered — it is not a No. If you represent Botpress, or anything here is wrong or out of date, you can supply it with a source: the operations this site accepts from an agent are listed at https://theagentsindex.com/.well-known/agents.json — that catalogue needs no key to read, and each entry states its own auth. Supplying evidence is free. Nothing that can be bought here changes what a field says, whether a gap is closed, or how this record is scored.