AI Tools
Decagon AI Agent Platform
Decagon AI Agent Platform: precios, capacidades, actualizaciones y alternativas, leídos en las páginas del propio desarrollador.

En inglés
Key facts
- Price
- Not stated on the maker’s pages
- Free plan
- Not stated on the maker’s pages
- Platforms
- Web
- API
- Yes Read on webflow2.decagon.ai
- Open source
- Not stated on the maker’s pages
- Last update
- 12 Nov 2026 Read on decagon.ai
Decagon helps customer support teams build agents that answer customer questions and carry out workflows across chat, voice, and email. Teams define agent behavior in natural-language procedures, connect business systems, and use testing and monitoring tools to refine performance.
The platform supports usage-based pricing per conversation or per resolved conversation, but Decagon’s pricing page returned a 404 and its blog lists no prices. Contact Decagon for a quote; the pages reviewed do not establish a free tier or a standalone knowledge-base product.
- Category
- Business & Productivity
- Read from the maker’s pages
- 2026-11-18
Company
Features
- Define agent behavior with natural-language Agent Operating Procedures (AOPs)
- Use Duet to analyze conversations and generate, test, and refine AOPs
- Connect agents to CRMs, help desks, knowledge bases, and call-center systems
- Set guardrails for workflows such as identity checks and refunds
- Test agents with simulated conversations and unit tests
- Version workflows and A/B test changes against real conversations
- Trace agent decisions, model calls, workflow triggers, and referenced articles
- Deploy agents across chat, voice, and email
Use cases
- Resolve customer questions across chat, voice, and email
- Automate identity checks and refund workflows
- Connect support agents to CRM and ticketing data
- Test and compare changes to customer-service workflows
- Coordinate multi-step customer outcomes across systems and people
Capabilities
- Writes and edits — “Decagon Assist autogenerates summaries, replies, and in-tool actions.” source
- Summaries — “Decagon Assist autogenerates summaries, replies, and in-tool actions.” source
- Automations — “Flexible, personalized chat that executes complex workflows reliably and empathetically.” source
- AI agents — “Rapidly build, optimize, and scale AI agents.” source
- Connects your apps — “Integrate with your full tech stack, including ticketing platforms, CRMs, knowledge bases, CCaaS providers, and other custom tools and systems.” source
- Admin controls — “Detection of personal agents and a dedicated channel to serve them, with permissions you control.” source
Latest updates
- Introducing Missions: Own end-to-end outcomes
Own end-to-end outcomes.
- Introducing Alto: the voice model driving a 5+ point lift in resolution
The voice model driving a 5+ point lift in resolution.
- How we taught our agents to recognize other agents
Taught our agents to recognize other agents.
- Hypothesis: Autopilot is better than ever. Results: Tested and validated.
Autopilot tested and validated.
- Introducing Experiment Hub: the latest experimentation platform on Decagon
Experiment Hub: experimentation platform on Decagon.
Where it works
Web
Alternatives to Decagon AI Agent Platform
- Decagon Voice
Decagon Voice handles customer support calls with AI agents that can resolve requests, take actions, and transfer calls to human agents.
- Agent Studio
Sierra Agent Studio lets teams build, test, and manage customer-facing AI agents using workflows, knowledge, and integrations.
- Ada AI Agent
Ada builds AI agents that answer customer questions and handle support tasks across channels, with tools for enterprise teams to manage them.
- Cresta AI Agent
Cresta AI Agent builds and runs customer-service agents that handle conversations, connect to business systems, and complete tasks.
- Zia Agents
Build and deploy AI agents that use business data and connected tools to carry out tasks and workflows.
- Crescendo CX Assistant
Crescendo’s AI agents handle customer support across channels, take actions, and pass unresolved issues to human specialists with context.
More from Decagon
- Decagon Voice
Decagon Voice handles customer support calls with AI agents that can resolve requests, take actions, and transfer calls to human agents.
Questions about Decagon AI Agent Platform
Does Decagon AI Agent Platform have an API?
Yes. The maker’s pages state that Decagon AI Agent Platform has an API. Read on the maker’s pages on 18 Nov 2026.
Which platforms does Decagon AI Agent Platform run on?
Decagon AI Agent Platform runs on Web, as its maker’s pages state. Read on the maker’s pages on 18 Nov 2026.
Is Decagon AI Agent Platform still available?
Yes, as of 18 Nov 2026: its maker’s pages were read that day and carried no notice that Decagon AI Agent Platform has ended.
What changed recently in Decagon AI Agent Platform?
The latest entry in its maker’s changelog: 12 Nov 2026, “Introducing Missions: Own end-to-end outcomes”. Before it: 11 Nov 2026, “Introducing Alto: the voice model driving a 5+ point lift in resolution”. Changelog checked on 16 Nov 2026.
Sources
- https://decagon.ai/product/overview — Product capabilities, workflows, testing, monitoring, and chat, voice, and email channels.
- https://decagon.ai/blog/pricing-ai-agents — Describes per-conversation and per-resolution billing; does not publish prices.
- https://webflow2.decagon.ai/product/integrations — Describes integrations, API and MCP support, and connections to support systems.
- https://decagon.ai/blog/missions — Announces Missions on 2026-11-12 and explains its multi-step workflow capabilities.
Not confirmed
- Prices
- The attempted official pricing page returned 404; no official pricing page with listed tiers or amounts was found.
- No tier prices were available on an official pricing page.
- Free plan
- The pages reviewed do not confirm whether a free tier exists.
- Open source
- The pages reviewed do not specify an open-source license or status.
- Platforms
- The pages reviewed describe agent channels, not supported client platforms.
- Latest version
- No current platform version or named release was identified.