GloriousFlywheel Enrollment Scoreboard Contract 2026-04-18

Snapshot date: 2026-04-18

Purpose

Define the reporting contract for orgwide enrollment so GloriousFlywheel status surfaces stop mixing together:

• real runner authority
• reusable-workflow consumption
• authority exceptions

This is the smallest note that turns the 2026-04-18 reality check into a stable scoreboard.

Why One Number Fails

One headline number hides three different truths:

• how many recent downstream repos actually run authoritative paths on GloriousFlywheel-backed runners
• how many repos merely consume shared templates
• how many repos only count if a non-default branch or non-primary repo is treated as authority

When those are collapsed, template usage and branch exceptions get mistaken for runner enrollment.

Scoreboard Rules

1. Use Default Branch As The Baseline

The normal orgwide scoreboard starts from the default branch.

That is the least gameable baseline and the best default for:

• current-state claims
• PM rollups
• issue and status updates

2. Track Authority Exceptions Separately

If a repo’s real product authority is not its default branch or not its primary non-fork repo surface, count that repo in a separate exception bucket.

Current example:

• Jesssullivan/MassageIthaca on dev/main

Do not silently add that repo into the default-branch denominator or numerator.

3. Template Consumption Is Not Runner Enrollment

A repo only counts toward runner enrollment if its authoritative workflow path visibly uses one of:

• an explicit shared ARC lane
• an explicit repo-owned runner contract

Template usage alone does not count.

4. Count Template Consumers By Mode

Reusable workflow consumers must be split into:

• hosted
• shared
• repo-owned

Only the last two can contribute to runner enrollment, and only when the authoritative workflow path really uses them.

5. Exclude Source And Template Repos From Downstream Adoption Claims

Use two denominators:

• workflow-surface context: all recent repos with workflows
• downstream adoption: recent workflow repos excluding tinyland-inc/GloriousFlywheel and tinyland-inc/ci-templates

This keeps platform self-proof and template infrastructure from inflating downstream adoption.

6. Do Not Count Raw Compatibility Lanes As Clean Enrollment

These do not count as clean runner authority unless they are the real intended contract on the authoritative path:

• literal [self-hosted, ...] compatibility jobs
• hardware-bound jobs
• one-off pet-runner lanes
• cache-only or Nix-only hosted flows

Required Weekly Scoreboard

Every orgwide status read should publish these numbers.

A. Workflow Surface Context

• recent non-fork repos
• recent repos with workflows
• recent downstream repos with workflows
• recent hosted-only workflow repos

B. Real Runner Enrollment

• recent downstream repos with shared ARC authority paths on default branch
• recent downstream repos with repo-owned authority paths on default branch
• authority exceptions with real runner authority off the primary counted surface
• downstream enrolled repos on default branch
• downstream enrolled repos with explicit branch-authority exceptions included
• broader cited-repo read with all named authority exceptions included

C. Template Consumer Modes

• hosted template consumers
• shared template consumers
• repo-owned template consumers

Current 2026-04-18 Snapshot

Workflow Surface Context

• recent non-fork repos: 88
• recent repos with workflows: 41
• recent downstream repos with workflows: 39
• recent hosted-only workflow repos on default branch: 20

Real Runner Enrollment

• shared ARC authority paths on default branch: 4
  • tinyland-inc/blahaj
  • tinyland-inc/lab
  • tinyland-inc/tinyland.dev
  • Jesssullivan/yt-text
• repo-owned authority paths on default branch: 2
  • Jesssullivan/scheduling-kit
  • Jesssullivan/acuity-middleware
• authority exceptions with real runner authority off the primary counted surface: 1
  • Jesssullivan/MassageIthaca on dev/main
• downstream enrolled repos on default branch: 6 / 39
  • about 15%
• downstream enrolled repos with explicit branch-authority exceptions included: 7 / 39
  • about 18%

Template Consumer Modes

• hosted template consumers: 6
  • tinyland-inc/tinyvectors
  • tinyland-inc/vite-plugin-a11y
  • tinyland-inc/tinyland-rate-limit
  • tinyland-inc/tinyland-physics
  • tinyland-inc/tinyland-color-utils
  • tinyland-inc/tinyland-a11y-engine
• shared template consumers: 0 proven on the scanned surface
• repo-owned template consumers: 2
  • Jesssullivan/scheduling-kit
  • Jesssullivan/acuity-middleware

PM Surface Wording

Use language like:

• downstream runner enrollment is 6 / 39 on default branches, or 7 / 39 when including one explicit branch-authority exception
• template adoption is 8 repos, but only 2 of those are repo-owned runner consumers and 6 remain hosted template consumers

Do not use language like:

• “8 repos use ci-templates, so 8 repos are enrolled”
• “recent org activity is mostly on GloriousFlywheel”
• “MassageIthaca is counted cleanly in the default-branch census”

Operational Use

Use this contract in:

• docs/current-state.md
• weekly PM and issue rollups
• adoption notes and queue updates
• Linear or status summaries where enrollment is discussed

If a repo changes mode, update the mode bucket first, then update the runner enrollment line only if the authoritative workflow path changed too.

Exit Condition

This note can retire once:

• default-branch and authority-branch reporting converge again
• broken self-hosted activation drift is gone from the cited repo set
• template mode is explicit for all active consumers
• orgwide status surfaces already report runner authority and template mode as separate numbers by default