[bokelley]

What

The storyboard matrices (storyboards, 14 jobs, and sales_storyboard_orchestrators, 8 jobs) now run on

runs-on: ${{ vars.STORYBOARD_RUNNER || 'ubuntu-latest' }}

The small sales_storyboards summary job stays on ubuntu-latest.

Why

The training-agent storyboard workflow uses about half of the repo's runner-minutes. During PR bursts it saturates the org's shared hosted pool of about 60 concurrent jobs (diagnosis in #8225). The org already has a 4-core larger runner, ubuntu-latest-m in the Default Larger Runners group. This change moves the matrices onto it without hard-coding it, so switching runners becomes a settings change, not a code change.

Rollout (settings, not code)

1. Raise the org's Actions spending limit. Larger runners are billed even for public repos, and ubuntu-latest-m is currently shut down at the cap.
2. In the Default Larger Runners group, grant adcp access and enable Allow public repositories.
3. Set the repository variable STORYBOARD_RUNNER=ubuntu-latest-m.

Until step 3, nothing changes. If the larger runner is ever shut down, for example by hitting the spending cap, delete the variable so jobs don't queue forever.

Verification

• tests/ci-runner-demand-workflows.test.mjs: 8/8, with a new test pinning the two heavy jobs to the variable and the fallback, and the summary job to ubuntu-latest.
• Changeset scope check passes. A patch changeset is included, because this workflow counts as protocol-scoped, as it did in fix(ci): skip storyboards on draft PRs and cancel superseded scans #8225.
• The pre-commit hook passed.

🤖 Generated with Claude Code

[aao-secretariat]

Ladon verdict: Escalate to human review

Escalate — gated path requires human/CODEOWNERS approval.

This PR modifies .github/workflows/training-agent-storyboards.yml, which matches the hard approval gate glob .github/workflows/** (gated_paths: true). The current review_decision is REVIEW_REQUIRED, not APPROVED, so the gate is not satisfied (decision table row 2).

The change itself is clean: it routes the storyboard and sales-orchestrator CI matrices through the STORYBOARD_RUNNER repo variable with a fail-safe ubuntu-latest fallback, preserves timeout-minutes: 30, keeps the summary job on ubuntu-latest, and ships a correctly-scoped patch changeset plus a test. The reviewer found no blocking, high, or medium findings.

Escalation reasons

• .github/workflows/training-agent-storyboards.yml is under a hard approval gate (.github/workflows/**) and review_decision is REVIEW_REQUIRED — human/CODEOWNERS approval required before merge.

Why human review

• .github/workflows/training-agent-storyboards.yml matches gated path .github/workflows/** and review_decision is REVIEW_REQUIRED (not APPROVED) — human/CODEOWNERS approval required.
• This PR touches a path under a hard, non-overridable approval gate (.github/workflows/training-agent-storyboards.yml (modified) matches .github/workflows/**) and the current GitHub review decision is 'REVIEW_REQUIRED', not APPROVED. This is a hard gate enforced in code — Ladon cannot auto-approve until a human/CODEOWNERS approval is recorded, regardless of how clean the diff is.