CARBON — Engineering ConfidenceUser Guide
CHAPTER 49 / 60 Pro

/carbon-generate

Generate prioritized CARBON test candidates from live pages, source code, requirements, screenshots, recordings, APIs, discovered widgets and controls, CARBON's built-in flow library, and user-defined or learned custom flows.

Jay · Your AI test managerTesting operations

When to use it

You need prioritized test candidates from a change or verified defect.

Before you start

Provide the requirement, failure evidence and missing coverage. Each proposal should have a discriminating assertion, setup and cleanup. Save candidates without calling them passed.

Try this in your coding agent.

/carbon-generate Propose regression tests for verified bugs in the password-reset flow.

Replace sample paths with your project. Slash names are shorthand; your host may use a plugin prefix or a skill picker. Invocation help →

ILLUSTRATIVE WALKTHROUGH

Password-reset regression candidates

An expanded teaching example, not a recorded or live execution. The report stays visible while the replay explains how to read it.

  1. 01Read context
  2. 02Analyze
  3. 03Show evidence
  4. 04Report

Coding-agent conversation

Read context

I’m reviewing password-reset regression candidates. I’ll keep the scope and permissions explicit.

Analyze

Analyze: Propose regression tests for verified bugs in the password-reset flow.

Show evidence

Here is the evidence trail to inspect: Candidate: expired token is rejected, password unchanged, helpful recovery link visible.

Report

A verified expiry defect becomes a concrete boundary-test proposal. Generated tests still need execution and review.

CARBON

Example report

Illustration
Inspect the identity boundary
Inspect the identity boundary — TestBucks teaching screen

TestBucks course screen with illustrative analysis. The screenshot supplies context; the demo does not establish a newly observed bug.

  1. User action
  2. Identity / validation
  3. Allowed outcome

Illustrated outcome

A verified expiry defect becomes a concrete boundary-test proposal.

Evidence to inspect

Candidate: expired token is rejected, password unchanged, helpful recovery link visible.

Still untested / not established

Generated tests still need execution and review.

Ready. Read the full example or play its steps.

What a useful result contains

A bounded claim

For password-reset regression candidates, state the target, revision, environment and actual work performed. Keep planned, executed, blocked and untested work separate.

An inspectable trail

Candidate: expired token is rejected, password unchanged, helpful recovery link visible. In a real run, retain the source, trace, screenshot or fixture behind the conclusion.

The qualification

Generated tests still need execution and review.

The next decision

Decide whether the evidence calls for a focused check, a reviewed fix or more context. Do not convert a plausible example into a verified finding.

Adapt the example to your project

  1. Replace password-reset regression candidates with a real feature, path or decision in your project.
  2. State the expected rule and its source. Include safe fixtures, constraints and a budget where appropriate.
  3. Ask for exact evidence and remaining gaps. Approve consequential actions separately.
/jay Review the /carbon-generate result. Separate observed evidence, assumptions and untested scope, then recommend the next useful check.

Where to go next

/carbon-tests →/carbon-test →

Content checked against the local command definition: skills/carbon-generate/SKILL.md. Examples adapted from the homepage’s TestBucks walkthroughs. Availability is edition- and host-dependent.