Flue
Flue is an open-source TypeScript framework for building durable AI agents with persistent state, multi-channel connectivity, and flexible deployment. Agents use a React-like hooks API to define behavior, maintain context across sessions, and automatically resume after interruptions. The framework supports deployment to Node.js, Cloudflare Workers, GitHub Actions, AWS, and containerized environments, and connects to communication platforms including Slack, Teams, Discord, and GitHub. Agents define tools and skills to call APIs, and can be orchestrated as workflows. Session traces export to observability platforms like OpenTelemetry and Braintrust for production monitoring and debugging.
Flue is an AI agent, integrating with Slack, Teams, Discord, and GitHub. InnovaAI scores it 4.6/10 for agency adoption, best for Backend Engineer, Technical Lead, and Operations Manager roles handling 5+ client meetings per week.
Agency Audit
Flue is an open-source TypeScript framework for building and deploying stateful AI agents with persistent memory, multi-channel connectivity, and LLM flexibility. Agencies building custom automation solutions, internal workflow orchestration, or client-facing agent products benefit most from adopting Flue internally. The framework's React-like hooks API, deployment flexibility across Node.js, Cloudflare, and GitHub Actions, and native integrations with Slack, Discord, Teams, and business tools like Stripe and Notion make it a strong fit for technical teams automating repetitive internal processes or prototyping agent-based client deliverables.
3recommended
36/mo
No paid plan published
Moderate
Illustrative scenario. Not a guarantee. Net capacity needs a verified paid base plan, and none is published for this service, so it is not modeled. Hours saved come from the service estimate; implementation, taxes, and unprovided usage charges are excluded.
- Backend Engineer handling multi-channel integration development
- Technical Lead handling long-running workflow automation
- Operations Manager handling state management for interrupted tasks
- Your team lacks TypeScript expertise and cannot allocate an engineer to learn the framework. Flue is not a visual builder; it requires code-first development.
- You need agents deployed and running within days, not weeks. Flue's learning curve and the need to define custom tools and skills mean initial agent development takes 2-4 weeks for a non-trivial workflow.
- Your workflows are simple enough for existing no-code automation tools like Zapier or Make. Flue's complexity overhead is only justified for multi-step, stateful, or highly customized agent behavior.
Internal Adoption Path
No paid plan published
36 hr/mo
3 seats × 12 hr each
$2,700/mo
modeled at $75/hr labor rate
No paid plan published
Illustrative scenario. Not a guarantee. No verified paid base plan is published for this service, so subscription cost and net capacity are not modeled. Implementation, taxes, and unprovided usage charges are excluded.
Platform Features
Core capabilities of Flue
React-like hooks API for agent behavior
Developers define agent logic using familiar hooks patterns, reducing the learning curve for teams already comfortable with React. This cuts the time for engineers to write their first production agent from weeks to days.
Persistent state and automatic resumption
Agents maintain context across sessions and resume interrupted work without manual intervention. Project managers and operations staff no longer need to manually restart failed workflows or re-enter context.
Multi-channel connectivity
Single agent codebase connects to Slack, Teams, Discord, GitHub, Twilio, WhatsApp, Telegram, and Google Chat. Technical operations teams eliminate the need to build separate integrations for each communication platform.
Flexible deployment across runtimes
Deploy agents to Node.js, Cloudflare Workers, GitHub Actions, AWS, Railway, Render, Fly, or Docker without code changes. Engineering teams avoid vendor lock-in and can migrate infrastructure without rewriting agent logic.
Tool and skill definition framework
Agents call APIs and perform actions through a declarative tool system. Developers spend less time writing boilerplate integration code and more time defining business logic.
Session trace export to observability platforms
Export traces to OpenTelemetry, Braintrust, or Sentry for debugging and monitoring. Technical leads gain visibility into agent decisions and can diagnose failures without digging through logs.
What Makes Flue Different
Unique advantages vs similar tools in this niche
Durable streams ensure agents never lose work and resume automatically after crashes
vs Stateless agent frameworks that lose context on failureFlue records every session in a durable stream and safely resumes interrupted work when the runtime comes back online.
React-like hooks API for building agents in TypeScript
vs Traditional imperative agent codingUse hooks like usePersistentState and useModel to manage state and model selection declaratively.
Deploy anywhere with support for Node.js, Cloudflare, and CI environments
vs Locked-in cloud-specific agent platformsWrite once and deploy to Node.js, Cloudflare, GitHub Actions, GitLab CI, and more.
Latest Updates
Recent releases and improvements for Flue
Flue 2.0, Introducing Agent Hooks
NewRelease of Flue 2.0 introducing Agent Hooks feature.
Value Equation
Outcome-likelihood-time-effort assessment for Flue
Value math requires real pricing
The Value Equation (dream outcome × likelihood ÷ time × effort) feeds directly into ROI math. Flue has no published pricing, so we hold this section until real numbers are available.
Contact FluePricing
Pricing data not yet available for Flue.
Reality Check
Flue requires TypeScript proficiency and assumes your team can manage deployment infrastructure or cloud platforms. The learning curve is steeper than no-code agent builders, and ROI depends on having workflows complex enough to justify custom agent development rather than off-the-shelf automation tools.
High effort: requires technical configuration and team training
How This Accelerates White-Label Services
Who It's For
- ✓ai-development-agencies
- ✓automation-consultancies
- ✓software-agencies-building-custom-agent-solutions
Acceleration Steps
- 1Schedule onboarding with the vendor
- 2Configure build durable ai agents with a react-like hooks api in typescript
- 3Connect Slack
- 4Launch your first client project
Academy for Flue
Work through it in order: the course for this service first, then the modules behind it.
No Academy modules are published for this service yet. Browse the full Academy
Why this category matters
The commercial case before the tooling.
Core concepts
The mental model you need to price and scope the work.
- Wiring Over WidgetsConcept
The AI agent itself is a commodity, but the value for agencies lies in the integration layer: connecting a pre-built agent to a client's CRM, calendar, and review cycle. This framework shifts focus from selecting the 'best' agent to mastering the wiring process. For example, an agency using Vendasta's white-label AI receptionist for a local business must configure it to match the client's booking rules and follow-up cadence, turning a generic tool into a tailored service. As agentic AI adoption grows (77% of decision-makers now run agents in production), clients expect this customization. Agencies that treat agents as components and invest in repeatable wiring processes can charge retainers for ongoing optimization, rather than one-off setup fees.
- Wiring Over WidgetsConcept
The AI agent market sells finished workers, but the strategic value for agencies lies not in the agent itself, which is increasingly a commodity, but in the wiring that connects it to a specific client's CRM, calendar, and review cycle. This framework, 'Wiring Over Widgets,' argues that agencies that treat agents as components rather than products win. The agent is the widget; the wiring is the integration, customization, and ongoing optimization that turns a generic tool into a tailored solution. For example, a white-label platform like Vendasta provides AI employees, but the agency's role is to configure them for each local business's unique lead flow and follow-up process. This wiring is where retainer pricing originates, as it requires ongoing maintenance and adjustment. Recent research shows that 88% of B2B marketers face foundational gaps, meaning clients need help not just deploying agents, but ensuring their operations can support them. Agencies that master the wiring can charge a premium for the irreducible value they add.
- Integration MoatConcept
The Integration Moat framework holds that the durability of an AI agent engagement is determined by how deeply the agent is wired into a client's existing systems, not by the agent's underlying capability. Since the agent itself is increasingly a commodity, the switching cost for the client lives in the integrations: the CRM fields mapped, the calendar sync, the review-cycle triggers, and the exception-handling rules. Agencies that invest in this wiring create a moat that competitors offering generic agents cannot cross. For example, a white-label platform like Vendasta lets an agency deploy an AI receptionist for a local business, but the real value is in configuring it to the client's booking flow and follow-up cadence. With 77% of AI decision-makers now running agentic AI in production, clients expect this depth, and agencies that deliver it convert one-off projects into retainers.
Decision and risk
How to judge the fit, and the ways it goes wrong.
- AI Agents Rule: Wire the Agent, Not the ProductEvaluation Rule
Treat the AI agent as a commodity component and focus your value on the integration into the client's specific workflows, systems, and review processes.
- AI Agents Rule: Wire the Agent, Not the ProductEvaluation Rule
Treat the AI agent as a commodity component and charge for the integration into the client's specific systems and workflows.
- The Productized Agent Trap: Why AI Agent Services Stall Without Client-Specific WiringFailure Pattern
- The Agent-as-Product Trap: Why AI Agent Services Stall Without Client-Specific WiringFailure Pattern
8 modules selected for Flue
Frequently Asked Questions
Answers about pricing, setup
Flue is an open-source TypeScript framework for building stateful AI agents that persist memory across sessions, connect to multiple communication channels like Slack and Discord, and deploy to any runtime including Node.js, Cloudflare, and GitHub Actions. Agents define tools and skills to call APIs and perform actions, and can be orchestrated as workflows. Session traces export to observability platforms like OpenTelemetry and Braintrust for production debugging.
Flue is open-source and free to use. There is no per-seat pricing, subscription fee, or commercial license required. Your only costs are infrastructure for deployment and any third-party LLM API calls your agents make.
Backend engineers and technical leads benefit most, as they write and maintain agent code. Operations and project management teams gain indirect value by automating repetitive workflows across Slack, Linear, Notion, and Stripe. Agencies building custom agent solutions for clients see the highest ROI, as Flue eliminates the need to build agent infrastructure from scratch for each project.
Time savings depend on your current workflow. Teams manually building multi-channel integrations or hand-coding state management for long-running tasks typically reclaim 4-8 hours per engineer per month once an agent is in production. Agencies building custom agents for clients compress agent development time by 30-40 percent compared to building integrations from scratch, translating to 6-12 hours saved per project.
Engineers familiar with TypeScript and React can write their first agent in 2-3 days. Teams new to agent development or TypeScript should budget 1-2 weeks for the first production agent. The hooks API and documentation reduce friction compared to building agents from first principles.
Yes. Flue is built on Pi, an agent harness that supports any LLM provider through its provider ecosystem. You can use OpenAI, Anthropic, Mistral, or any other LLM without changing agent code.