Run SWEny triage to investigate production alerts, create issues, and optionally implement fixes. Use when the user wants to analyze recent errors or alerts.

2026-04-063

workflow-diagram

swenyai/sweny

Render a SWEny workflow as a Mermaid diagram for visualization.

2026-04-063

workflow-edit

swenyai/sweny

Edit an existing SWEny workflow using natural language instructions.

2026-04-063

name	e2e-run
description	Run SWEny E2E browser test workflows from .sweny/e2e/. Use when the user wants to execute end-to-end tests.
disable-model-invocation	true
allowed-tools	Bash
context	fork
agent	sweny-workflow
argument-hint	[file] [--timeout <ms>]

SWEny E2E Run

Run agent-driven browser end-to-end tests. Executes workflow files from .sweny/e2e/.

Prerequisites: E2E workflows must exist in .sweny/e2e/. Run /sweny:e2e-init first if the directory is empty.

Usage

Run all E2E tests:

sweny e2e run

Run a specific test file:

sweny e2e run $ARGUMENTS

With a custom timeout (default is 15 minutes per workflow):

sweny e2e run --timeout 300000

What to expect

Each test workflow takes 1-15 minutes depending on flow complexity
Tests use agent-browser for accessibility-tree-based browser automation (not screenshots)
On completion, report: pass/fail status for each flow, any errors encountered
Test data matching e2e-{run_id} patterns is automatically cleaned up if a cleanup backend is configured

Environment variables

E2E_BASE_URL — override the base URL for tests
E2E_EMAIL — override test email (default: e2e-{run_id}@yourapp.test)
E2E_PASSWORD — override test password