all

wiki/lint-reports/2026-07-18-lint-full-wiki.md

@@ -0,0 +1,66 @@
+# Lint Report — Full Wiki (2026-07-18)
+
+#lint-report
+
+## Scope
+
+Full-wiki health check of all **40 pages** (1 overview · 13 sources · 17 concepts · 6 entities · 1 comparison · 1 query · 1 lint report), the follow-up to the sources-only lint ([[2026-07-18-lint]]). Every concept, entity, comparison, and query page was read in full; link health was checked against a deduped inventory of all `[[…]]` targets across the vault.
+
+Checks: page-type tags · template-section completeness · broken links · orphans / weak linking · cross-page contradiction coverage · stale claims superseded by newer sources · missing evidence links · high-mention concepts lacking pages.
+
+## Verdict
+
+**The wiki is in good structural health.** Tags, templates, links, and orphan-coverage all pass cleanly. The only real findings are **staleness introduced by today's adversarial-source ingest** — three pages still say "the vault has no adversarial source," which is no longer true. Two fixes + one clarification applied; one structural gap (a missing *referrals* page) flagged.
+
+## Passing Checks
+
+- **Page-type tags: 100%.** All 40 pages lead with the correct folder-matching type tag (`#source`/`#concept`/`#entity`/`#comparison`/`#query`/`#overview`/`#lint-report`), topical tags after. No mismatch.
+- **Template sections: complete.** All 17 concepts and 6 entities carry the required Summary / Current Understanding / Evidence / Related Pages / Contradictions-Uncertainty / Next Questions. The query matches the Query template; sources verified in the prior lint.
+- **No broken links.** Every `[[…]]` target resolves to an existing page. (Two `[[…]]` hits are illustrative text inside code spans; one `[[wiki/overview|Overview]]` is a path-style link — resolvable but inconsistent, fixed below.)
+- **No orphans; linking is strong.** Every page has ≥7 inbound references (busiest: [[client-acquisition-channels]] 46, [[methodology-as-moat]] 43, [[2026-07-17-design-the-perfect-offer]] 42; lowest are the query at 7 and this-day's lint at 4 — both expected for their type). The new [[ai-productivity-evidence]] already has 26 references — well-integrated on arrival.
+- **Evidence links present** on every concept/entity page.
+- **Contradiction coverage is thorough.** Every live tension is recorded on the relevant page(s): productize-immediately-vs-after-2–3 ([[productized-service]]), one-package-vs-price-ladder ([[offer-ladder]]), cloning-vs-moat ([[cloning-over-originality]]), online-vs-in-person ([[client-acquisition-channels]]), two-moats ([[methodology-as-moat]]/[[relationships-as-moat]]), and now the foundational AI-capability contradiction ([[ai-productivity-evidence]]). No unrecorded contradiction found.
+
+## Findings
+
+### F1 — `productized-service` still says the vault has no adversarial source (Medium) — ✅ fixed
+
+The page's Contradictions and Next Questions read "still no true adversarial view" / "the vault needs an adversarial source," now contradicted by the day's ingest of [[ai-productivity-evidence]]. It also leans on the "~$200/mo AI substitute" premise that the new evidence contests.
+**Fix:** updated both sections to acknowledge the adversarial source exists and to sharpen the *remaining* gap — the new source tests the **AI-capability premise**, not the **productization model itself**, so a *productization-failure* source is still genuinely missing. Added a "now contested" pointer on the substitution bullet and [[ai-productivity-evidence]] to Related Pages.
+
+### F2 — Query snapshot's "no adversarial source" caveat is now imprecise (Low) — ✅ fixed (dated note)
+
+[[2026-07-17-best-method-first-client]] states "the vault also has no adversarial or failure-case source on any of this." True for its *channel/sales* topic, but the blanket phrasing is now misleading.
+**Fix:** appended a dated lint note (preserving the original per the Update Policy) clarifying that the new adversarial source addresses the AI-capability premise, not these channel claims — which remain un-countered.
+
+### F3 — Inconsistent path-style link in `index.md` (Low) — ✅ fixed
+
+`index.md` used `[[wiki/overview|Overview]]` (path-style) where the vault everywhere else uses the bare `[[overview]]`. Resolvable but inconsistent. **Fix:** changed to `[[overview|Overview]]`.
+
+### F4 — High-mention concept with no page: **referrals** (Medium) — flagged
+
+Referrals appear as a distinct, high-value channel with their own statistics and mechanics across [[2026-06-15-17-ways-first-client]] (91% would refer / 11% asked; highest-converting; ~0 cost; ask-immediately-after-delivery), [[sales-discipline]], [[client-acquisition-channels]], and the query — but have **no dedicated concept page**. This is the clearest structural gap. Recommend creating `wiki/concepts/referrals.md` (inbound from client-acquisition-channels, sales-discipline; the "can't bootstrap client #1" sequencing constraint is its key nuance). Not created in this lint pass (a lint reports; it doesn't author content).
+
+### F5 — Secondary concept candidate: **guarantee / risk-reversal** (Low) — flagged
+
+The guarantee mechanic recurs with an explicit open question across [[pricing-from-value]] (the guarantee test) and [[outcome-based-selling]] ("guarantee mechanics are undefined"). Currently well-covered within those two pages; a dedicated page is optional, lower priority than referrals.
+
+### F6 — Optional enrichment links to the new evidence page (Low) — flagged
+
+[[niche-selection]] ("'go where AI is powerless' assumes LLM capability is static") and [[solution-vs-staff-augmentation]] ("directly substituted by the ~$200/mo alternative") both make claims the new [[ai-productivity-evidence]] speaks to. Not errors — optional cross-links that would tighten the web. Left for a future pass to avoid over-editing during a lint.
+
+## Note — a nuance worth recording (not a defect)
+
+The adversarial ingest was appropriately *surgical*: it added counter-evidence only to the three pages it genuinely contradicts ([[ai-market-shift]], [[seniority-and-ai]], [[future-of-engineering-work]]). It did **not** touch [[relationships-as-moat]] or [[product-ownership]], which rest on the "AI levels skill" premise — correctly, because "novices gain most" *corroborates* leveling rather than contradicting it. So the same source contradicts one Sebastian claim (juniors irrelevant) while supporting another (skill is leveled); the vault currently records the contradiction but not the corroboration. Optional future touch.
+
+## Changes Made This Pass
+
+- [[productized-service]] — F1: adversarial-source acknowledgement + substitution-premise pointer + Related-Pages link.
+- [[2026-07-17-best-method-first-client]] — F2: dated clarifying note.
+- `index.md` — F3: path-style link normalized (plus this report added to the Lint Reports table and counts updated).
+
+## Recommended Next Operations
+
+1. **Create the `referrals` concept page** (F4) — the one real structural gap.
+2. **Ingest a productization-failure-case source** — still the highest-value content gap (the AI-capability premise now has a counter; the *selling model* does not).
+3. Optional tidy: F5 (guarantee page), F6 (enrichment links), the corroboration note above.