Agent Studio · early access

Creating Agents
Doesn't Have To Be Hard

The management layer your team needs to build, run and benchmark agents in production. Across the apps your team already uses.

Go To Dashboard

Join the community

Servflow

✦ Assist

Agents

Conversations

Traces

Resources

Integrations

Providers

Secrets

Workspaces

Describe the agent you want to build

◷ Previous conversations 1

Ask Assist to build an agent

Describe an agent, or ask about the one you have

Suggested

A Telegram bot that answers questions about my docs→

Review pull requests on GitHub when they open→

A daily digest of a Discord channel, every weekday at 9→

An HTTP endpoint that triages support tickets→

Review pull requests on GitHub when they open

✦ Here's an agent that reviews every new pull request.

PR reviewer Apply

pr-reviewer · 1 agent node · 1 entry · 0 context groups

To fill in the studio

agents[0].providerPick the model the agent runs on

⚡GitHub webhook

✦ Sub-agent

PR reviewer

anthropic · claude-sonnet-5-5

2 tools · 1 entry

GitHub Fetch

GitHub Post

⚡ pull_request · opened

⌥ diff · 3 files

servflow-bot reviewed · just now

Looks good. One fix: the error from parseConfig is dropped at config/load.go:42.

pr-reviewerrun 7f3a9cPull request opened4.2 sOK

0 s1 s2 s3 s4 s

Agent run PR reviewer

4.21 s

Model call claude-sonnet-5-5

1,240 tok

Tool GitHub Fetch

diff · 3 files

Model call claude-sonnet-5-5

2,572 tok

Tool GitHub Post

comment posted

Model

claude-sonnet-5-5

Input tokens

3,120

Output tokens

692

Cost

$0.021per run

What is Servflow?

Agent designer and runtime.

Design your agent visually. Lay out its prompts, tools, decision branches and guardrails on a single canvas, then run it on the same platform you built it in.

Add a sub-agent, attach its tools, pick its model. See the whole architecture at a glance instead of tracing it through code.

Adding a sub-agent to an agent on the Servflow canvas

Visual canvasBranching logicActionModel-agnosticGit syncMCP

Built to handle complexity

Empowering technical and non-technical agent builders.

Product and ops teams design agents on the canvas. Engineers open the same agent as YAML to review it in a pull request, version it in Git, and ship it with the rest of the code.

No code

Drag in sub-agents, wire tools, pick models

Code

YAML config, Git history, one CLI command to create

name: Support triage
config:
  entries:
    - type: handler
      handler: github
      start: agent.classifier
  agents:
    classifier:
      providerID: 1
      systemPrompt: Label the issue.
      next: agent.researcher
    researcher:
      providerID: 2
      tools: [web_search, http, github_mcp]
      next: agent.responder
    responder:
      providerID: 1
      next: end

The Support triage agent on the visual canvas

Canvassupport-triage.yaml

Built to handle security

Built for security and production-grade load.

Agents run shell commands, read files and call tools. Servflow runs all of it inside an operating-system sandbox, on an engine written in Go so it stays small and quick.

Anthropic-grade sandboxing

Memory efficient, single Go binary

Fast, concurrent runs

sandbox · bwrap / seatbelt

shell · run 7f3a

$ git clone repo && make test

own folder · /runs/7f3a

Networkoff by default

~/.ssh, /etchost files hidden

Server processnot visible

one Go binaryconcurrent

run 1

run 2

run 3

run 4

run 5

secrets

{{ secret "github-token" }}

••••••••••resolved at run time

One-click runsActionLive test outputInstant re-runsAction

Tool

100s of services and actions, a tool call away

Sign in to the apps your team already uses and their actions show up as tools, next to ServFlow's own. Point a sub-agent at an MCP server, or hand a task to another sub-agent on the canvas.

Apps