Jank Reports — Admin

Self-test results

End-to-end smoke run against bing.com.

Run a test

Paste one or more URLs — each is crawled, audited, and queued as a report.

URLs Intensity

default

Standard 1 credit / page Main + 2 sub-pages · personas, flows, basic WCAG · ~5 min

3× depth

Deep 3 credits / page + Exploratory agent · VPAT WCAG · voice narration · ~12 min

wide

Superwide 1.5 credits / page Crawls ~every one-hop page + exercises its actions · default 25 pages

deep+

Super-deep 4 credits / page Top business-critical flows, executed 10–15 steps deep · default 15

all-in

Complete 5 credits / page Wide AND deep — every one-hop page, top flows run deep, VPAT WCAG · can take hours

tune

Custom Live formula Every knob exposed below · 50-credit cap / submit

—

Coverage

tested vs. theoretical max

—

Work

est. wall-clock

—

Value

~ manual-QA equiv.

—

Cost

page-credits

Advanced settings visibility · brand · personas · flows · subpages · WCAG · exploratory · auth

Visibility

Brand (report theme)

Label (optional)

LLM Provider

LLM Model

Tunnel (optional)

Email report links to (optional, comma-separated)

Competitive runs — AI finds competitors of the first URL and runs the same audit on them (public). Reuses any already-tested.

Min competitors

Max competitors

Competitor type / prompt (optional)

Specific competitors (optional, one per line or comma-separated)

Test flows (recommended)

Generate & replay AI-proposed user flows; only passing flows surface in the report

How many flows?

Custom flow context (optional)

Persona feedback (optional)

Generate diverse user personas and collect their feedback on the page

How many personas?

Custom persona context (optional)

Subpages (additional pages tested)

Also audit AI-picked subpages from the entry URL (signup, pricing, checkout, docs, etc.)

How many subpages?

Accessibility audit (recommended)

Run rule-based WCAG audit (A / AA / AAA grades + legal & user-impact analysis)

Audit depth

Exploratory agent (recommended)

Run an AI exploratory tester that takes interactive actions on the page (click / type / submit) and reports findings

Steps

Mode

Type

Focus

Custom instructions (optional)

Advanced — headers & auth (optional)

Add custom HTTP headers, basic / bearer / cookie auth, or override the User-Agent (useful for bypassing bot detection or auth-walled pages)

curl -X POST /api/reports -H "Authorization: Bearer $TOKEN" -d '{"urls":["…"]}'

Internal staff with bypassed daily caps. Manage via ADMIN_EMAILS env var on Cloud Run (semicolon-separated).

Email	Role	Has Customer Record	Runs	Last Run

Public pages describing each test type — what it covers, what it finds, what it doesn't, and a sample finding from the report schema. Use these for prospect calls, customer-success replies, or anywhere you need to point someone at "what does Jank actually test?"

Adding a new test type? Drop the HTML in public/test-types/, add a server route in server/index.js (search for /test-types/), and add an entry to TEST_TYPE_DOCS below.

Designed for the large monitor in the office. All three render the world map exactly once (no horizontal wrap), surface the live queue + the run currently in flight, and animate every ~1.2 s. Use the floating pill at the top of the wallboard to switch layouts, or press 1 2 3.

P1Mission Control

NASA / Bloomberg crossover — calm, dense, monospace numerics. Best for a wallboard that runs for hours.

P2Neon Grid

Cyberpunk / Tron — pure black, neon green & magenta, scan-lines. Loudest visual; great for glance-and-go.

P3Tactical HUD

Aerospace / military — amber + electric blue, hex motifs, radial gauges. Maximalist; every pixel populated.

P4Marquee

Clean, giant — 5 huge stats across the top with count-up tweens, then a 4×3 grid of real screenshots cycling through each app's runs.

Each row turns a one-off report submit into a recurring run on the chosen cadence. Cloud Scheduler hits POST /api/admin/schedules/tick every 5 min and queues a fresh report for every schedule whose nextRunAt has passed. After each run, the owner gets the standard report email + a "what changed since last run" diff with a one-click stop link.

Loading…

Global kill-switches. Existing accounts and in-flight reports are NOT affected — these only block new activity. Every change emails jason@testers.ai (override via FEATURE_FLAG_NOTIFY_TO env).

Conversion = the account reached a real paid Stripe plan (has upgradedAt / firstPaidPlan). Accounts merely assigned a plan by an admin (no Stripe subscription) are counted as not converted.

Acquisition → paid, by source

Loading…

Nurture emails sent

Loading…

Email	Company	Plan	Status	Runs	Limits (perDay · maxPages)	Created

Email	Company	Plan	Status	Runs	Limits (perDay · maxPages)	Created

Chatbot red-team test

Point at any page with a chatbot. We find it, probe it (positive · negative · edge · jailbreak · misuse · vertical), and score its strengths, weaknesses & risks.

Page URL Context & expectations — optional, steers what we test & how we judge

Depth

Email

Reports / day

Max URLs / request

Note

Email	Key	/day	Max URLs	Active	Created	Note

Admin sign-in required

Mission Control

Lifetime impact …

Self-test results

Run a test

Acquisition → paid, by source

Nurture emails sent

Chatbot red-team test

🌍 Live submission map loading…

Runs / day loading…

New emails / day

Issues found / day

Admin sign-in required

Mission Control

Lifetime impact …

Self-test results

Run a test

Acquisition → paid, by source

Nurture emails sent

Chatbot red-team test

🌍 Live submission map loading… All Flagged Blocked Demo

Runs / day loading…

New emails / day

Issues found / day

🌍 Live submission map loading…