test: golden valuation fixture shared with recoupable/marketing

[sweetmantech]

Part of recoupable/chat#1850 (video/experience parity, item 2 — valuation derivation guard).

Why

The catalog valuation model is intentionally implemented twice: lib/catalog/computeValuationBand.ts here (server-side valuations) and computeCatalogValuation.ts in recoupable/marketing (the /valuation card computes its band client-side for the progressive in-card state). The constants are mirrored by comment-discipline only — nothing fails if one side drifts. This PR adds the guard.

What

• lib/catalog/__tests__/fixtures/valuation-golden.json — canonical inputs → expected band cases derived independently from the published formula ($0.0035/stream, gross-ups 1.25/1.4/1.6, NLS = gross x 0.85 x 0.75, multiples 10/13/16, age = round(ms/365.25d) clamped >= 1, default 5y). Cases: the real verified 16.3B-stream catalog (age fallback), zero streams, a 10y catalog aged from its release date, and the 1y age clamp.
• lib/catalog/__tests__/valuationGoldenFixture.test.ts — runs computeValuationBand against every case, asserting to within one cent. GREEN on first run against the existing implementation (zero divergence today).

Lockstep rule: this fixture's twin lives at marketing/lib/valuation/__tests__/fixtures/valuation-golden.json and the two files must stay byte-identical (SHA-1 a2c9737f167a25e3589b5d53fa4bf832d4f1e0bd). Any model change must update the fixture in BOTH repos in the same coordinated change, or one repo's CI goes red — which is the point.

Follow-up (out of scope): extract a true shared package so the formula exists once; until then this fixture is the divergence tripwire.

Verification

• pnpm test — 714 files / 3,944 tests pass (4 new)
• pnpm exec tsc --noEmit — 200 errors, identical to the pre-existing main baseline (none in touched files)

Siblings

• Marketing twin: test: golden valuation fixture shared with recoupable/api marketing#44
• Chat starter-template PR: feat: prefer the Weekly valuation + streams report starter template chat#1856

Note: branch is tests/valuation-golden-fixture (not test/...) because this repo's long-lived test branch occupies that ref namespace.

🤖 Generated with Claude Code

---

Summary by cubic

Adds a shared “golden” valuation-band fixture and test to guard against drift between the server computeValuationBand and marketing’s client implementation. The byte-identical JSON is committed in both repos and CI asserts each case to the cent.

• Migration
  • If the valuation formula or constants change, update the fixture in both repos in the same change.

^{Written for commit 69ac021. Summary will update on new commits.}

[cursor]

Bugbot is not enabled for your account, so this pull request was not reviewed.

Enable Bugbot in the Cursor dashboard to get automatic reviews on future PRs.

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
api	Ready	Preview	Jul 7, 2026 11:23pm

[coderabbitai]

Important

Review skipped

Review was skipped due to path filters

⛔ Files ignored due to path filters (2)

• lib/catalog/__tests__/fixtures/valuation-golden.json is excluded by !**/__tests__/** and included by lib/**
• lib/catalog/__tests__/valuationGoldenFixture.test.ts is excluded by !**/*.test.*, !**/__tests__/** and included by lib/**

CodeRabbit blocks several paths by default. You can override this behavior by explicitly including those paths in the path filters. For example, including **/dist/** will override the default block on the dist directory, by removing the pattern from both the lists.

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

Run ID: 29215e0e-d22d-48b4-b4e3-883cbd7ad2a2

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch tests/valuation-golden-fixture

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[cubic-dev-ai]

No issues found across 2 files

Confidence score: 5/5

• Automated review surfaced no issues in the provided summaries.
• No files require special attention.

Architecture diagram

sequenceDiagram
    participant Test as Test Runner
    participant Fixture as valuation-golden.json
    participant Compute as computeValuationBand()
    participant MarketFixture as Marketing Twin Fixture
    participant CI as CI Pipelines

    Note over Test,Compute: NEW: Cross-repo divergence guard
    Test->>Fixture: Load cases from JSON
    loop For each case
        Test->>Compute: computeValuationBand(input)
        Compute-->>Test: { valuation, catalogAgeYears }
        Test->>Test: Compare catalogAgeYears
        Test->>Test: Compare low, mid, high (≤0.01)
    end
    alt All cases match
        Test-->>CI: Test passes
    else Any mismatch
        Test-->>CI: Test fails → divergence detected
        Note over CI,MarketFixture: Must update fixture in BOTH repos
    end
    Note over Fixture,MarketFixture: Byte-identical (SHA-1 check)

Loading

<sub>Auto-approved: Adds golden test fixture and test for valuation band calculation; no production code changed.

Re-trigger cubic</sub>