frontend-e2e

A live end-to-end test command for a frontend web application, meaning the part of a website that users see and interact with. It uses a real browser to check pages and user flows.

In plain words
What is it for?
Use it to test navigation, forms, controls, responsive layouts, accessibility, visual behaviour, performance, and page-level errors.
Why use it?
It checks the complete site rather than only isolated functions, and continues testing other pages when one check fails.

Command

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/intellegix/intellegix-code-agent-toolkit/frontend-e2e
Clone the repo
git clone --depth 1 https://github.com/intellegix/intellegix-code-agent-toolkit
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 15,617 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00000 $0.15617
Opus 5 $0.00000 $0.07809
Sonnet 5 $0.00000 $0.03123
Haiku 4.5 $0.00000 $0.01562

Measured 3d ago against content hash 8a2dc5c5ee97, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

frontend-e2e scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

commands/frontend-e2e.md · 1,175 lines

How it starts

The opening of the file, as written. The whole thing — 1,175 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/frontend-e2e — Comprehensive Frontend E2E Live Test

Perform a thorough live E2E test of a frontend web application using browser-bridge MCP tools. Tests every page, interactive element, form, navigation flow, responsive breakpoint, and accessibility rule through real browser interaction.

Architecture: Detect -> Crawl -> Universal -> A11y -> Visual/UX -> Navigation -> Responsive -> Interactive -> Performance -> Report. $0 cost, fully local via browser-bridge.

CRITICAL: Execute the full test silently. Only interact with the user in Step 0 (URL/env confirmation) and Step 0.5 (auth probe). Present the full report at the end.

IMPORTANT: Per-page error isolation is MANDATORY. Every page's test suite is wrapped in error handling — if any test fails or errors, mark it FAIL and continue to the next page/test. NEVER abort the full run because of a single page failure.

Input

$ARGUMENTS = URL [flags]

Flag Description Default
--focus <path> Test only pages matching this path prefix all pages
--skip-a11y Skip accessibility checks (Tier 1.5) enabled
--skip-responsive Skip responsive/breakpoint tests (Tier 3) enabled
--max-pages <n> Override default page cap 20
--deep Set page cap to 50 instead of 20 off
--auth-cookie <name=value> Inject a session cookie before crawl none
--skip-visual Skip visual/UX quality checks (Tier 1.7) enabled
--skip-design Skip visual design quality checks (Tier 1.8) enabled
--allow-submit Allow form submission (localhost/staging only) off

Step 0: Detect Application — MANDATORY, SILENT

  1. Parse $ARGUMENTS for URL and flags
  2. If no URL: probe common dev server ports (3000, 3001, 4200, 5173, 5174, 8000, 8080) via mcp__browser-bridge__browser_navigate to http://localhost:{port} — use first that responds
  3. If still no URL: STOP — "Provide a URL or start a local dev server"
  4. Call mcp__browser-bridge__browser_get_tabs to verify browser-bridge connection. If fails: STOP — "Browser-bridge not connected. Check Chrome extension."
  5. Production safety gate: If URL is NOT localhost, 127.0.0.1, or does NOT contain staging/dev/test in hostname:
    • Present warning: "This appears to be a production URL. Interactive tests (Tier 4) will NOT submit forms. Use --allow-submit only for test/staging environments."
    • Set internal flag SAFE_MODE = true (no form submissions, no button clicks that trigger navigation away)
  6. Navigate to URL via mcp__browser-bridge__browser_navigate
  7. Wait for page: try mcp__browser-bridge__browser_wait_for_element targeting body (timeout 15s). Known issue: wait_for_element may return "restricted URL" on localhost — if so, fall back to browser_evaluate with !!document.body check + 2s delay as workaround.
  8. Take baseline screenshot via mcp__browser-bridge__browser_screenshot

Read the full file on GitHub · 1,175 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 3d ago First seen · 1,175 lines · 0 tokens per session scan A 8a2dc5c5ee97

Subscribe to this mod's changes

frontend-e2e is a command published in the GitHub repository intellegix/intellegix-code-agent-toolkit (57 stars, last pushed 9d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 15,617 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.