Getting started

This guide walks you from zero to a published flow you can call over HTTP. It covers both the web console path and the CLI + git path.

Prerequisites

• Email address for signup at fetchcatch.com
• (Optional) fcc CLI for repo sync
• (Optional) An OpenAPI JSON URL from an API you want to call from flows

Path A — Console first (recommended for new users)

Step 1: Create an account

1. Go to fetchcatch.com/signup
2. Create your tenant account
3. You land in the console with your workspace ready

Step 2: Add an API source

1. Open API sources in the sidebar
2. Click Add source
3. Paste your OpenAPI document URL (e.g. https://api.example.com/swagger/v1/swagger.json)
4. Optionally select an auth profile if the API requires credentials
5. Save — operations appear in the flow designer palette

→ Details: API sources

Step 3: Create a response type (optional but recommended)

1. Open Response types
2. Add fields your flow should return (e.g. approved: boolean, reason: string)
3. Mark required fields

→ Details: Response types

Step 4: Design a flow

1. Open Flows → New flow
2. Drag nodes from the palette: Start → HTTP → Condition → Decision → End
3. Wire edges; configure JSONata on conditions
4. On the start node, bind your response type
5. Save the draft

→ Details: Flow designer

Step 5: Publish

Publish from the designer or flows list. Only published versions are callable via /v1/evaluate.

Step 6: Create an API key

1. Open API keys
2. Create a key with evaluate permission
3. Copy the secret — shown once

→ Details: API keys

Step 7: Call your flow

curl -X POST "https://api.fetchcatch.com/v1/evaluate/your-flow-slug" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"userId": "u-123", "amount": 100}'

→ Details: Evaluate API

Path B — CLI + git (recommended for teams)

Step 1: Install the CLI

Download from fetchcatch.com/downloads, then:

fcc install
fcc version --check

Step 2: Initialize your repo

From your project root:

fcc init --workspace my-project

Creates .fetchcatch/ including AGENTS.md for AI coding tools.

Step 3: Log in

fcc login

Opens the browser for email/password sign-in. Credentials are stored in ~/.fetchcatch/, not in your repo.

Step 4: Pull existing flows

If you already built flows in the console:

fcc pull

Files appear under .fetchcatch/flows/ and .fetchcatch/response-types/.

Step 5: Edit and sync

# edit JSON locally or in Cursor
fcc status          # see what changed
fcc apply --dry-run # preview push
fcc apply           # push drafts to server
fcc publish         # ship callable version for /v1/evaluate

Step 6: Automate with CI

Add a GitHub Actions workflow to run fcc apply then fcc publish on merge to main.

→ Example: Sync & CI

The day-to-day loop

Design in console  ──pull──►  Edit in git  ──apply──►  Live on server
       ▲                                              │
       └──────────── tweak in console ◄────────────────┘

Troubleshooting

Next steps

• Core concepts — deeper mental model
• Flow designer — visual authoring
• CLI reference — all commands
• Runs & debugging — inspect live executions