Botpress
Open-source conversational AI platform for building and deploying chat agents with LLM integration, NLU, and a visual flow builder. Botpress v12 is self-hosted and open source; Botpress Cloud is the managed offering. Provides a TypeScript SDK for building custom actions and integrations, a visual conversation flow editor, and built-in NLU for intent classification. Widely used for customer service bots, internal automation, and LLM-powered chat agents.
Score Breakdown
⚙ Agent Friendliness
🔒 Security
SOC2 Type II for cloud. GDPR compliant. Self-hosted option for full data sovereignty. AGPLv3 for auditability. Webhook signature verification available. HTTPS enforced for cloud.
⚡ Reliability
Best When
Teams wanting an open-source, self-hostable chat agent platform with LLM integration, visual flow editor, and TypeScript extensibility — especially if avoiding vendor lock-in is important.
Avoid When
You need a fully autonomous agent loop without conversation flow constraints, or your team is Python-primary — code-first frameworks give more flexibility.
Use Cases
- • Build LLM-powered customer service agents with structured fallback flows for when AI doesn't have a confident answer
- • Create autonomous AI agents in Botpress with custom TypeScript actions that call external APIs and business logic
- • Deploy chat agents to multiple channels (WhatsApp, Telegram, Slack, web) from a single Botpress project
- • Build self-hosted chat agents with full data control using Botpress open source — no cloud dependency
- • Create knowledge base agents using Botpress's built-in document knowledge base (RAG) without external vector DB setup
Not For
- • Python-first teams — Botpress's custom action SDK is TypeScript/Node.js; Python agents are better served by LangChain or Griptape
- • Real-time voice agents — Botpress is text-first; Vapi, Bland.ai, or Retell are better for voice
- • Highly complex agent loops requiring custom orchestration — Botpress's flow model constrains agent logic compared to code-first frameworks
Interface
Authentication
Personal access tokens for API and SDK access. Bot-level tokens for channel integrations. Self-hosted version uses internal auth tokens. Botpress Cloud uses account-level PATs. Webhook signatures for securing incoming webhooks.
Pricing
Open source (AGPLv3) is free for self-hosting. Cloud free tier is limited for production. Paid cloud plans based on message volume and AI credits. Self-hosting avoids SaaS costs but requires DevOps.
Agent Metadata
Known Gotchas
- ⚠ Botpress Cloud vs self-hosted have different capabilities — some cloud features (AI agents, built-in LLM) are cloud-only; self-hosted requires more manual LLM integration
- ⚠ Custom actions in Botpress use TypeScript with the @botpress/sdk — Python teams must wrap their logic in a Node.js action layer
- ⚠ Knowledge base RAG quality depends on document upload and configuration in the visual editor — no programmatic control of RAG index
- ⚠ AGPLv3 license requires open-sourcing derivative works in some commercial contexts — verify license compatibility before building a commercial product on Botpress
- ⚠ Conversation state is stored per-session in Botpress — long conversations may accumulate large context; implement summarization for very long chats
- ⚠ Integration with external channels (WhatsApp, Telegram) requires configuring platform-specific webhooks in both Botpress and the channel provider
Alternatives
Full Evaluation Report
Detailed scoring breakdown, competitive positioning, security analysis, and improvement recommendations for Botpress.
Scores are editorial opinions as of 2026-03-06.