liveBook · Manning

Green classification

As explained in Vibe Engineering

A change-risk classification that permits a pull request to be merged automatically when all required safety conditions are met.

Green classification marks a change as safe enough to merge automatically when its policy checks pass. Safety depends on context, not diff size alone: a 30-line change in a high-impact core module may be Yellow, while the same change in a low-risk notification leaf may be Green. section 10.5.2

Define Green with explicit, enforceable conditions:

  • Require measurable checks such as test or coverage results, maximum diff size, and restrictions on new dependencies. section 10.5.2
  • Exclude high-risk surfaces or changes, including schema changes, authentication modifications, external API changes, and money or user-data paths. section 10.5.2
  • Route Yellow changes to an appropriate owner and Red changes to senior or security review. section 10.5.2
  • Start conservatively, measure unmodified approvals, promote reliable patterns, and recalibrate monthly. section 10.5.3

Policy-based classification

A core library can be Green only above 95% coverage, below a 50-line diff, and without new dependencies. Checkout requires passing tests with no schema or authentication changes. Notification permits passing tests, no external API changes, and a diff below 100 lines. section 10.5.2

Listing 10.4 Policy-based ownership with triage rules
libs/core:
  owner: platform-team
  green: "coverage > 95%, diff < 50 lines, no new dependencies"
  yellow: "platform-team-oncall reviews"
  red: "platform-team-lead + security-review"
 
services/checkout:
  owner: payments-team
  green: "tests pass, no schema changes, no auth modifications"
  yellow: "payments-team-member reviews"
  red: "payments-team-lead reviews"
  sla: "99.99% - minimum Yellow review for all changes"
 
services/notification:
  owner: platform-team
  green: "tests pass, no external API changes, diff < 100 lines"
  yellow: "any platform-team member reviews"
  red: "platform-team-lead reviews"

Inspect the policy listing to compare module-specific Green conditions with Yellow and Red review ownership. section 10.5.2

Treat every Green merge as a risk-and-capacity bet. Automation is worthwhile only when escape costs remain below avoided review costs; released review capacity must then fund product work or controls that reduce future queue size. section 10.5.3

How Vibe Engineering covers it

How deep it goes
Defines module-level criteria and calibration practices, but stops short of specifying implementation machinery.
Examples
Uses core-library, checkout, and notification policies with coverage, diff, dependency, schema, authentication, and API conditions.
What it assumes
Assumes observable tests, coverage, diffs, dependencies, ownership, and module risk surfaces.
What you can do after
Operate/configure policy-based auto-merge, measure escapes, and adjust Green criteria using observed approval accuracy.
How it connects
Connects auto-merge with code review, ownership, blast radius, testing, coverage, and risk-based escalation.

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Vibe Engineering. The section links point to the parts of the book it draws on.