Fully autonomous, steerable web page testing and QA.
Give CARBON a URL. It explores your site, exercises real journeys, and brings back the bugs—with evidence. Set the direction. Let your AI QA sub-agents do the work.
USD · New plans are arranged with our team. Execution allowance, supported checks, and spending cap are agreed before activation. No checkout or automatic billing here.
Start with a guided Cloud pilot
$299 · 30 days · one-time fee
One product, an agreed set of important journeys, and a findings review. Testing allowance and human support are scoped before you start. No automatic renewal.
CloudThe same AI QA team. Cloud or local. Local
Runs in your coding agent
CARBON Local Harness
Your coding agent. Now with a test team.
One command. An entire verification loop.
Specialist AI QA sub-agents, right where you build. CARBON reads your code and existing tests, follows the risk, tests your app, and explains what’s ready—and what isn’t.
In your coding agent · Your tools · You stay in control
Inside your AI coding agentIllustrative chat
You
/carbon
A preview of the CARBON workflow.
01Understand your code & coverageFind the gaps that matter.
02Explore. Test. Follow the evidence.Real interactions. Specialist perspectives.
03Explain your release confidenceKnow what to fix and test next.
Simulated example. No tests are running on this page.
/Explore all CARBON commandsOne testing loop. 55 ways to go deeper.
Start with /carbon, or focus on the area that matters to you. Add your feature, URL, or question after a command. Names and invocation syntax can vary by coding agent.
55 commands
Start here & keep testing
/carbon
Run the complete risk-based testing loop, from code and coverage to evidence and confidence.
/carbon-help
Find the most useful testing workflow for your project.
/carbon-demo
Try CARBON on a fresh, disposable sample project.
/carbon-test
Focus testing on one feature, flow, page, or behavior.
/carbon-more
Run the next highest-value checks left from an earlier pass.
/carbon-auto
Run a bounded test, fix, and retest loop.
/carbon-forever
Keep exploring and expanding coverage until stopped or limited; requires two confirmations.
/carbon-issues
Hunt reproducible bugs through code inspection, stateful journeys, and persona exploration.
/carbon-browser
Exercise a real page or user journey in an approved browser.
Explore, plan & understand
/carbon-map
Explore a visual product map of pages, widgets, journeys, risks, and test coverage.
/carbon-personas
Get goal-driven AI persona feedback on your product, document, or concept.
/carbon-confidence
Investigate consequential questions and reassess release confidence as evidence arrives.
/carbon-human
Capture the human decisions and business context that testing cannot infer.
/carbon-feedback
Collect and retain scoped feedback locally or through your own reviewer service.
/carbon-full-report
See the complete quality record, automation, findings, outcomes, and trends.
/carbon-stats
Review commands, tests, findings, and measured usage where available.
Focused quality checks
/carbon-accessibility
Check accessible interactions and WCAG criteria; produce evidence-backed reports and draft ACRs.
/carbon-security
Investigate security weaknesses, access boundaries, and unsafe behavior.
/carbon-privacy
Check personal-data exposure, storage, consent, retention, and deletion.
/carbon-functionality
Verify that features behave as intended, including edge and failure cases.
/carbon-ui
Find visual, layout, and interface consistency problems.
/carbon-ux
Evaluate end-to-end experience, expectations, and journey friction.
/carbon-usability
Check whether people can understand and complete their tasks.
/carbon-performance
Investigate slow responses, rendering delays, and performance bottlenecks.
/carbon-reliability
Exercise error handling, recovery, and dependable behavior.
/carbon-compatibility
Check behavior across relevant browsers, devices, and environments.
/carbon-content
Review clarity, accuracy, consistency, and product messaging.
/carbon-data
Check data integrity, transformations, persistence, and round trips.
/carbon-api
Test API contracts, validation, permissions, and error responses.
/carbon-networking
Investigate connection failures, retries, timeouts, and network behavior.
/carbon-localization
Review translations and locale-specific behavior in product context.
/carbon-intl
Assess internationalization readiness, formats, text expansion, and locale support.
/carbon-agentic
Assess how easily an AI agent can understand and safely use your page or API.
/carbon-geo
Review discoverability, grounded content, and readiness for AI-search answers.
/carbon-state
Hunt lost, stale, or inconsistent state across tabs, sessions, and data round trips.
/carbon-load
Assess load readiness and optionally run explicitly approved, bounded traffic tests.
/carbon-stress
Investigate stress and recovery limits; non-local execution requires two confirmations.
/carbon-failure
Inject failures into isolated copies to test error handling without changing originals.
Build tests & connect tools
/carbon-generate
Generate prioritized test candidates from code, pages, requirements, and evidence.
/carbon-tests
Create, import, organize, repair, and manage reusable tests.
/carbon-ai-upgrade
Modernize existing tests with AI-assisted checks and workflows.
/carbon-chatbot
Build or extend a broad chatbot test suite and optionally exercise it.
/carbon-integrations
Connect issue and test-management systems; external writes require explicit approval.
/carbon-settings
Set project context, browser controls, credentials, and report preferences locally.
AI evaluations & release decisions
/carbon-confidence-init
Initialize a project’s confidence model, risk record, and evidence structure.
/carbon-confidence-plan
Plan risk-based validation for an AI feature, model, prompt, or agent change.
/carbon-eval
Design and run evaluations for models, prompts, RAG, ranking, and agents.
/carbon-eval-design-review
Challenge whether an evaluation measures the right things with representative cases.
/carbon-review
Independently review AI-generated code, behavior changes, and supporting evidence.
/carbon-security-review
Review AI-specific threats such as prompt injection, leakage, and cross-tenant access.
/carbon-skeptical-review
Challenge assumptions, optimistic claims, and gaps in release evidence.
/carbon-statistical-review
Review evaluation statistics, uncertainty, sampling, and comparisons.
/carbon-release
Recommend ship, canary, hold, or rollback based on evidence; humans retain release authority.
/carbon-release-review
Independently synthesize release risks, severe failures, and rollback readiness.
/carbon-incident
Contain an AI incident, preserve traces, investigate causes, and add regression checks.
No matching commands. Try another topic or clear your search.
Testing specialties are not paid unlocks. Run CARBON in your existing tools; your agent subscription or API usage is separate. Open-source release planned.
Optional human services · separate from the software
USD · Deliverables, included effort, response expectations, and any additional costs are agreed before work begins. No unlimited consulting or automatic subscription.
One team. Cloud or local.
Meet your AI QA sub-agents.
Different perspectives. One clearer picture of quality. Select a profile to see what each specialist looks for.
AI perspectives, not human reviewers. Coverage depends on your target and the checks performed.
AI
AI QA sub-agent
AI generated code
Deliberately hunts for the bugs AI-generated code can introduce and the gaps AI coding agents can overlook. Goes beyond the happy path: missing boundary and null checks, invented API assumptions, incomplete error handling, async races, state lost across multi-step journeys, privacy leaks, and tests that pass without checking the intended behavior. Challenges plausible-looking code with targeted failure cases and evidence.
What I look for
boundary and null cases
invented API assumptions
error and recovery paths
async races
multi step state loss
privacy leaks
weak test assertions
For example
Save a draft, sign out, then switch accounts. Does the previous user’s draft appear?
An AI QA sub-agent, not a human reviewer. Actual coverage depends on your target, access, and the checks performed.
WCAG specialist focused on criterion-level evidence across perceivable, operable, understandable, and robust behavior, without overstating automated scan results as conformance.
What I look for
accessibility
a11y
wcag
conformance
For example
Increase text size to 200%. Can you still read the price and reach the checkout button?
An AI QA sub-agent, not a human reviewer. Actual coverage depends on your target, access, and the checks performed.
Compatibility specialist focused on browser, device, viewport, operating-system, assistive-technology, and support-matrix evidence, with explicit coverage gaps rather than assumed portability.
What I look for
compatibility
cross browser
responsive
devices
assistive technology
For example
Open checkout on a narrow phone screen. Does the keyboard cover the payment action?
An AI QA sub-agent, not a human reviewer. Actual coverage depends on your target, access, and the checks performed.
Performance specialist focused on user-visible latency, Core Web Vitals, payload and request cost, caching, API timing, scalability, mobile constraints, memory, and leaks.
What I look for
performance
perf
web vitals
networking
reliability
For example
Filter a long product list on a slow connection. Does the interface stay responsive?
An AI QA sub-agent, not a human reviewer. Actual coverage depends on your target, access, and the checks performed.
Content and UX-writing specialist focused on page identity, clear copy, information architecture, credibility, navigation, status communication, readability, and content quality.
What I look for
content
copy
spelling
seo
ui
For example
Compare the offer headline with its checkout terms. Do they promise the same thing?
An AI QA sub-agent, not a human reviewer. Actual coverage depends on your target, access, and the checks performed.
Forms specialist focused on input contracts, validation, boundaries, state, submission, recovery, data quality, conversion barriers, and accessible interaction.
What I look for
forms
form
functionality
data
For example
Submit a form with whitespace-only required fields. Does validation explain what to fix?
An AI QA sub-agent, not a human reviewer. Actual coverage depends on your target, access, and the checks performed.
First-impression and conversion specialist focused on value clarity, trust, navigation, calls to action, responsive composition, dead ends, and page credibility.
What I look for
landing page
landing
ui
ux
For example
Follow the primary call to action. Does the next screen deliver what the headline promised?
An AI QA sub-agent, not a human reviewer. Actual coverage depends on your target, access, and the checks performed.
Checkout and payment specialist focused on order accuracy, address and payment input, trust, retry safety, pricing truth, completion, and conversion-blocking failures.
What I look for
checkout
ecommerce
payment
functionality
For example
Retry a checkout after a timeout in a test environment. Could the order be submitted twice?
An AI QA sub-agent, not a human reviewer. Actual coverage depends on your target, access, and the checks performed.
Pricing and subscription specialist focused on plan clarity, comparison, currency and locale, hidden conditions, billing cadence, conversion paths, and truthful claims.
What I look for
pricing
localization
For example
Switch between monthly and annual billing. Are the billing interval and total charge explicit?
An AI QA sub-agent, not a human reviewer. Actual coverage depends on your target, access, and the checks performed.
