Introduction
Learn what Failpath is, how the CLI and SDK fit together, and the key concepts you need to get started.
Quick Start
Initialize Failpath in your repo, instrument a backend function, and publish your first flow in minutes.
CLI Reference
Explore the
init, publish, and sync commands that manage your flow graphs.SDK Reference
Browse the full
@failpath/sdk API, including typed flow keys, run(), step(), and more.Get started
Follow these four steps to go from a blank repo to a live, monitored flow on the Failpath dashboard.1
Create a project in the Failpath dashboard
Sign in to failpath.dev and create a new project. Once it’s created, copy your project key — it looks like
fp_project_xxx and is the only credential you need to connect your repository.2
Initialize your repository
Run the following command at the root of your repo, replacing This creates a
fp_project_xxx with your actual project key:.env file with FAILPATH_PROJECT_KEY, a .failpath/flows.json file containing your dashboard’s current flow graph, a .failpath/sdk.ts helper for typed flow keys, a .failpath/AGENTS.md file with step-key guidance, and updates .gitignore to keep secrets out of version control.3
Install @failpath/sdk and instrument your code
Install the SDK in your backend project:Then wrap your backend functions with the generated typed helper,
run(), and step() so Failpath can track each step as it executes:4
Publish your flow to the dashboard
Push your local flow graph to Failpath so the dashboard reflects your code:Failpath validates
.failpath/flows.json before pushing, so any structural issues are caught before they reach the dashboard.