Log inSubmit server
WalkerOS Flow logo

WalkerOS Flow

Build, validate, load, and simulate walkerOS event flows locally through MCP.

InstallRepositoryWebsite
Full README Compare alternatives Embed widget Claim this listing

Install WalkerOS Flow

Pick your client, then add it in one click or paste the config.

STDIONode.js
~/Library/Application Support/Claude/claude_desktop_config.json
claude_desktop_config.json
{
  "mcpServers": {
    "walkeros-flow": {
      "command": "npx",
      "args": [
        "-y",
        "walkeros"
      ]
    }
  }
}

Settings → Developer → Edit Config, merge this into mcpServers, then fully quit and reopen Claude Desktop.

Install not yet sandbox-tested— we haven't run this command through our automated sandbox yet. Not a red flag; we're working through the catalog.

Overview

walkerOS Flow MCP server exposes walkerOS event-pipeline workflows to AI agents over a local Model Context Protocol interface. It can load flows, validate configurations, and simulate events without an account, while walkerOS also supports schema inspection, tracking-definition suggestions, and integration-code generation. Reach for it when you want an agent to work with versioned event collection configuration before connecting browser or server sources to destinations such as GA4, BigQuery, or ad platforms.

Use cases

  • Load a walkerOS flow for agent-assisted review
  • Validate event-pipeline configuration before deployment
  • Simulate events against a local flow
  • Inspect event schemas and suggest tracking definitions
  • Generate integration code from tracking requirements

Key features

  • Local flow loading
  • Flow validation
  • Event simulation
  • Event schema inspection
  • Tracking-definition suggestions
  • Integration-code generation

Security & health

What it can reach and how reliably it runs — from our own checks, not the maintainer’s claims.

ActiveLast checked 1d ago

Recent health check succeeded.

8/8 checks healthy over the last 45d

Supply-chain signal

No high-severity advisories surfaced by our automated scan.

Critical 0High 0Medium 0Low 0

Scanned 8d ago via OSV.dev · walkeros (npm)

Adoption & maintenance

Factual signals from GitHub, npm, and our automated checks — not a rating.

GitHub stars347Stargazers on the source repository.
npm downloads2.3kPackage downloads in the last 30 days.
Last commit1d agoMost recent push to the default branch.

How WalkerOS Flow works

What the walkerOS Flow MCP server does

The walkerOS Flow MCP server gives MCP-compatible AI agents access to walkerOS event-pipeline workflows. Its documented local workflow includes loading a flow, checking whether its configuration is valid, and simulating an event. These operations let a developer test collection behavior before wiring a production source or destination.

walkerOS is an event collection layer built around sources, a collector, and destinations. Sources can include browser code, data layers, Express applications, AWS Lambda, and GCP Functions. The collector handles processing such as consent, validation, mapping, routing, and enrichment. Destinations include GA4, Google Ads, Meta CAPI, BigQuery, and other supported packages.

The MCP interface also supports agent-assisted work with event schemas. Agents can read a schema, suggest tracking definitions, and generate integration code.

How it works

The walkerOS Flow MCP server runs locally through the @walkeros/mcp package. An MCP client starts it with npx, then communicates with the process using the client’s MCP configuration. Flow loading, validation, and event simulation do not require a walkerOS account.

A flow can be used in either an integrated or bundled walkerOS setup. Integrated mode imports startFlow into a TypeScript application. Bundled mode uses a JSON flow definition and the npx walkeros bundle flow.json command to produce a standalone script.

Setup and configuration

For clients other than Claude Code, add a server entry that starts @walkeros/mcp with npx:

config.json
{
  "mcpServers": {
    "walkeros-flow": {
      "command": "npx",
      "args": ["@walkeros/mcp"]
    }
  }
}

Claude Code can install the walkerOS plugin from the project marketplace with /plugin marketplace add elbwalker/walkerOS, followed by /plugin install walkeros@elbwalker. The repository also documents a separate browser-source MCP server, @walkeros/mcp-source-browser; that component is not the flow server itself.

Tools and capabilities

The documented capabilities include:

  • Load walkerOS flow configuration locally.
  • Validate flow definitions and their configuration.
  • Simulate events against a flow.
  • Read event schemas for agent-assisted tracking work.
  • Suggest tracking definitions.
  • Generate integration code.

The broader walkerOS project supports configuration as code, declarative tagging, consent-aware event queuing, schema validation, routing, and multiple destinations. Whether a specific capability is available through the flow MCP process should be checked against the current MCP documentation.

Limitations and notes

It also does not document environment variables for this server. Local flow operations are explicitly described as requiring no account, but destination credentials or external service configuration may still be needed when implementing a real pipeline.

The repository is MIT licensed. Documentation for the MCP application is available at the walkerOS documentation site under the MCP section.

Read the full README View source on GitHub

Reviews

No reviews yet — be the first to share how it worked for you.

Frequently Asked Questions about WalkerOS Flow

walkerOS Flow is an MCP server that connects AI agents to walkerOS event pipeline workflows through local flow loading, validation, and event simulation tools. It can also support reading event schemas, suggesting tracking definitions, and generating integration code. It runs without an account.

Install it by adding an MCP server entry that runs `npx @walkeros/mcp`. For Claude Code, use the walkerOS marketplace and plugin installation commands documented in the repository. Other MCP clients can use the provided `command` and `args` configuration.

No, loading a flow, validating it, and simulating an event run locally without an account. Separate credentials or service configuration may still be necessary for real destination integrations.

It can let AI agents read event schemas, suggest tracking definitions, and generate integration code.

Related MCP servers

All in Developer Tools All alternatives
Add a badge to your READMEShow it’s listed — live SVG, Markdown or HTML.

AllMCPs Directory Badge

Full Badge Customizer

Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.

Badge Style:
Live Dynamic SVG PreviewWalkerOS Flow AllMCPs Directory Badge
Markdown (GitHub README)
[![AllMCPs](https://allmcps.com/api/badge/walkeros-flow?style=directory)](https://allmcps.com/mcp/walkeros-flow)
HTML Embed
<a href="https://allmcps.com/mcp/walkeros-flow"><img src="https://allmcps.com/api/badge/walkeros-flow?style=directory" alt="WalkerOS Flow on AllMCPs" /></a>