/carbon-release
Decide whether an AI system or AI-generated change should ship, canary, hold, or roll back using traceable evidence, uncertainty, severe failures, monitoring, and reversibility.
Jay · Your AI test managerProof & release confidenceWhen to use it
You need a scoped recommendation to ship, canary, hold or roll back.
Before you start
Bring current evidence, severe unresolved failures, monitoring and reversibility. The recommendation should state blockers and uncertainty; humans retain deployment authority.
Try this in your coding agent.
/carbon-release Assess whether the support assistant is ready to ship; list blockers and uncertainty.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
Support-assistant release gate
An expanded teaching example, not a recorded or live execution. The report stays visible while the replay explains how to read it.
- 01Read context
- 02Analyze
- 03Show evidence
- 04Report
Coding-agent conversation
I’m reviewing support-assistant release gate. I’ll keep the scope and permissions explicit.
Analyze: Assess whether the support assistant is ready to ship; list blockers and uncertainty.
Here is the evidence trail to inspect: Map blocking findings to release criteria, owner actions and remaining uncertainty.
The example recommendation is hold until refund-policy failures are resolved. The demo makes no real release decision.
Example report
Illustration
TestBucks course screen with illustrative analysis. The screenshot supplies context; the demo does not establish a newly observed bug.
- Question / intent
- Reference evidence
- Scoring & uncertainty
Illustrated outcome
The example recommendation is hold until refund-policy failures are resolved.
Evidence to inspect
Map blocking findings to release criteria, owner actions and remaining uncertainty.
Still untested / not established
The demo makes no real release decision.
Ready. Read the full example or play its steps.
What a useful result contains
A bounded claim
For support-assistant release gate, state the target, revision, environment and actual work performed. Keep planned, executed, blocked and untested work separate.
An inspectable trail
Map blocking findings to release criteria, owner actions and remaining uncertainty. In a real run, retain the source, trace, screenshot or fixture behind the conclusion.
The qualification
The demo makes no real release decision.
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
- Replace support-assistant release gate with a real feature, path or decision in your project.
- State the expected rule and its source. Include safe fixtures, constraints and a budget where appropriate.
- Ask for exact evidence and remaining gaps. Approve consequential actions separately.
/jay Review the /carbon-release result. Separate observed evidence, assumptions and untested scope, then recommend the next useful check.Where to go next
/carbon-release-review →/carbon-full-report →Content checked against the local command definition: skills/carbon-release/SKILL.md. Examples adapted from the homepage’s TestBucks walkthroughs. Availability is edition- and host-dependent.