Flow Runner

Chain multiple HTTP requests into automated workflows with variable passing, lifecycle scripts, and execution metrics.

What is Flow Runner?

Flow Runner lets you build multi-step HTTP workflows where each step can extract values from its response and pass them to subsequent steps. Instead of manually copying tokens, IDs, or other values between requests, Flow Runner automates the entire chain.

Key Concepts

  • •Flow Group: A collection of related flows, like a folder. Each group can have its own environments (dev, staging, prod).
  • •Flow: A sequence of HTTP request steps that run together. A flow belongs to exactly one group.
  • •Step: A single HTTP request within a flow. Each step has its own URL, method, headers, body, authentication, assertions, and variable extractions.

Hierarchy

Flow Group (e.g., "Auth API")

→ Flow (e.g., "Login Flow")

→ Step 1: POST /api/login

→ Step 2: GET /api/profile

→ Step 3: PUT /api/settings

→ Flow (e.g., "Registration Flow")

→ Step 1: POST /api/register

→ Step 2: POST /api/verify-email

Use Cases

API Test Automation

Chain login → create resource → verify → cleanup. Run with assertions to validate your API contracts.

Regression Testing

Run flows with multiple iterations to catch intermittent failures and performance regressions.

Request Chaining

Extract tokens, IDs, and URLs from responses and inject them into subsequent requests automatically.

Lifecycle Setup & Reporting

Use Before All to fetch a shared auth token once, then use After All to ship JSON/HTML/PDF reports to your CI, dashboard, or chat webhook.

Features

  • 1Lifecycle scripts: Before All / After All scripts run once per flow, plus per-step pre and post scripts for fine-grained control
  • 2Variable extraction: Pull values from responses using JSON path, headers, regex, or status code
  • 3Iterations & parallel execution: Run sequentially N times for stability testing, or N concurrent virtual users for load-style runs
  • 4Execution metrics: Duration, success rate, average response time, data transfer, and bottleneck detection
  • 5Anomaly detection: Multi-iteration runs flag outlier and inconsistent steps based on response time variance
  • 6Reporting: Export run results as HTML, PDF, or OTEL JSON — manually from the UI or programmatically via at.export
  • 7Import steps: Import requests from Box (saved requests) or Projects with a single click
  • 8Flow environments: Group-level environment variables with multiple environments (dev, staging, prod) and OS keychain–backed secrets
Box (Saved Requests)The Interface