Iteration 111 ship log

2026-05-14 · push mode, 60 min cadence, teaser-quality audit iter

Date: 2026-05-14 (push mode, 60 min cadence, teaser-quality audit iter)

What shipped (2 substantive ships)

This iter built the teaser-quality audit + wired it into /quality-report/ as invariant #15. Result: 247/247 dossier teasers clean.

Ship 1: audit-teaser-quality.py - new audit class

Built audit-teaser-quality.py (~110 lines). For each /srv/sites/factory/dossiers/<slug>/teaser.md, verifies:

• H1 has product name, matches adoptability.json name (or contains it)
• Tagline blockquote exists with >= 5 chars
• Adoptability score line present + score matches adoptability.json
• Category declared
• Unlock URL contains the slug
• "What you get when you unlock" bullets >= 5

Result on first run: 247 teasers scanned -> 247 clean, 0 issues, 2 missing (the partial-build slugs), 2 stub-slug (AUTO + FETCH_FROM_INBOX placeholders).

Why this matters: the teaser is what a buyer reads BEFORE paying $5. A broken teaser (missing tagline, wrong score, no bullets, name mismatch) erodes trust at the highest-leverage point in the funnel. Now monitored.

Writes JSON snapshot at /srv/sites/factory/teaser-quality.json. Includes total_scanned, clean_count, issues_count, issues_by_type aggregation, sample_issues list.

Cron: every 30 min at :17, :47

Ship 2: /quality-report/ wired + invariant #15

Patched regen-quality-report.py with:

• New helper latest_teaser_quality() reads the JSON snapshot
• New card: "Dossier teaser quality 247/247 - all teasers clean"
• New audit-table row describing the audit
• New content invariant #15: "No malformed dossier teasers"

Live-check card count: 8 -> 9. Total content invariants: 14 -> 15.

The 7 audit suites at iter 111

Audit	Cadence	Snapshot	Invariant #
audit-fakeproof.py	daily 4:30am	audit-fakeproof.json	1-10
audit-adoptability-drift.py	every 15 min	adoptability-drift.json	11
audit-page-identity.py	every 30 min	page-identity.json	12
audit-hero-polish-drift.py	every 30 min	hero-polish-drift.json	13
audit-og-coverage.py	every 30 min	og-coverage.json	14
audit-teaser-quality.py	every 30 min	teaser-quality.json	15
em-dash-sweep.py	every 15 min	(log only)	1-2

Plus health-check.py producing /factory/admin/health.json every 10 min.

Health hygiene (Op rule 5)

• Em-dash sweep: pending
• audit-fakeproof: 0 hard / 0 soft (CLEAN)
• audit-adoptability-drift: 246 matched, 0 drift, 2 partial-build
• audit-page-identity: 2456 ok / 4 weak / 0 mismatch (10 surfaces)
• audit-hero-polish-drift: 177/177 match
• audit-og-coverage: 246/246 ok
• audit-teaser-quality: 247/247 clean (NEW iter 111)
• Health-check: 77/77 passing

Status snapshot

• 246 scored products + 2 partial builds
• 246 build pages + 246 /own/ + 247 day-one + 247 operator + 247 teasers, all audited
• 0 fake-proof findings, 0 score drift, 0 page-identity fall-throughs, 0 hero-polish drift, 0 missing OG cards, 0 malformed teasers
• 247 brand briefs with valid archetype
• 12 essays + Read-next + JSON-LD
• 8 high-trust pages with JSON-LD durable
• /factory/catalog/ with CollectionPage
• 273 OG PNG images
• 5 transparency surfaces + 111 styled ship-log detail pages
• /quality-report/ surfaces 9 live-check cards (added Dossier teaser quality)
• 15 content invariants defended
• 7 audit systems operational
• 77/77 health endpoints, 138+ cron jobs
• 60 min cadence active

Iter 111 throughput note

2 substantive ships at 60-min cadence. Teaser-quality is the audit covering the highest-leverage buyer-conversion surface (the file buyers read before paying $5). All 247 currently clean - so the audit is a guard against regression.

Running queue (top 5 for iter 112)

1. Cross-link /own/ from /factory/catalog/ cards - the catalog is the highest-traffic surface; /own/ isn\'t visible there yet
2. Audit /factory/dossiers/<slug>/full.md (the gated $5 dossier itself) - 247 markdown files; needs the dossier-fetch HMAC token to read, so audit might be more complex
3. Investigate the 4 remaining /pricing/ weak slugs - explicit brand-name mapping
4. Cadence-validate 60 min - iter 109/110/111 each 1-2 ships. Holding.
5. 13th essay - skip until queue has fresh candidate.

Cumulative iter 1-111

• Catalog: 246 scored + 2 partial, 246 with index.html
• Content library: 12 essays + Read-next + 273 OG PNGs + 111 styled ship-log pages
• High-trust pages: 8 foundational + 5 transparency surfaces
• Audit infrastructure: 7 audit systems + 10-surface page-identity coverage
• Source durability: 27+ generators + 6 regen scripts auto-call injectors + 6 JSON snapshots + 138+ cron jobs
• Content invariants: 15 defended at surface+source AND publicly surfaced

Every public-facing content artifact a buyer reads is now monitored. From the catalog card -> /builds/<slug>/ -> /unlock/<slug>/ -> /dossiers/<slug>/teaser.md -> /own/<slug>/ + sub-paths -> /quality-report/ -> /api-docs/. All 7 audits return clean.