@clize/clize 0.35.4 → 0.35.5

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@clize/clize",
-  "version": "0.35.4",
+  "version": "0.35.5",
   "mcpName": "ai.clize/clize",
   "description": "The real-world capability layer for AI agents — domains, email, deploy, media, SEO/GEO. CLI + MCP.",
   "keywords": [

package/skills/clize-seo/SKILL.md

@@ -43,8 +43,8 @@ on our own tenant zero, 2026-09): every funded row is a head term — `stretch`
 `no_volume` / `pre_emergence` cell. Eleven weeks, 45 pages, 0 clicks, and the only cluster with
 impressions was one Google picked for us. The read is `keywords.byBand` against the ledger's
 `format` column: if the ledger holds no row your authority can win, placement is not the fix
-either — re-derive the ledger toward tool, long-tail and locale cells (§1.2, §7.2) before
-pitching a roundup. The counter-evidence sits in the same dogfood: tabledi and kunavo earned
+either — re-derive the ledger toward tool, long-tail and locale cells (§1.2, §7.2), building
+the tools if the site has none, before pitching a roundup. The counter-evidence sits in the same dogfood: tabledi and kunavo earned
 page-1 positions with zero links, on tool and long-tail pages. Settle it per site with one round
 of such rows enrolled in `check`, not by argument.
 
@@ -117,7 +117,7 @@ the cross-product. Mechanical is the point.
 | Competitor displacement | each competitor brand | `alternative` / `vs <you>` / `pricing` / `review` | `openrouter alternative` |
 | New-thing window | model / API / spec names released in the last 90 days | `api`, `pricing`, `how to use` | `seedance api pricing` |
 | Host × capability | each agent host you have actually tested | each thing the product does for it | `openclaw email`, `codex deploy site` |
-| Tool × locale | each browser-side tool you can honestly ship | each market you can serve | `hreflang generator` in `ja-JP` |
+| Tool × locale | each browser-side tool you can honestly ship — built, or not built yet (§7.2, deriving the tool set) | each market you can serve | `hreflang generator` in `ja-JP` |
 | Locale × head term | your two or three head terms | each market you can serve | `de-DE` pricing cluster (price it; §7.2 on whether to build it) |
 
 - **Write every cell, then filter.** Filtering while enumerating is recalling again.
@@ -573,7 +573,8 @@ a family generator existed, and tabledi shipped forty in two days from one. Buil
 before the second page of any family, and never machine-translate a family page.
 
 **On a site that already has tools, this round's page count is the locale siblings of the tools
-it has**, and the gate is each cell's own SERP (§7.2). Siblings come in two prices. A *format*
+it has**, and the gate is each cell's own SERP (§7.2); a site without tools derives the set
+first (§7.2). Siblings come in two prices. A *format*
 tool (viewer, merge, split, convert) is one strings entry (tabledi: a `DICT` table in
 `tools/apps.js`; clize.ai: a `strings.<loc>.mjs`) — ten in a session is ordinary. A *text* tool
 (duplicates, compare, clean) carries locale semantics — full-width and half-width forms in
@@ -611,6 +612,21 @@ intent row → use-case; tool-intent row → tool page.
 ### 7.2 Tool pages
 
 The one page type where the content *is* the function — and the strongest link asset (§6.1).
+
+**A site with fewer than five tools has no multiplier, and inventing the tool set is its first
+production task** — not a product decision to wait for. Three sources, all paid for already or
+free: (1) the sweep (§1.1) — every rival keyword shaped like a calculator, converter, counter,
+checker, estimator or generator; tabledi's teardown found rows.com, formulabot and quadratic
+earn their organic traffic from free calculators, not product pages, and that finding sat in a
+ledger for three days without becoming a rule; (2) the product's own data — a catalog with
+prices is a cost calculator, a limits table is a context-window checker, a media catalog is a
+generation-cost estimator, a list of formats is a converter; every number the product already
+maintains is a tool nobody else can keep current; (3) the neighbours of the tools you have.
+Each candidate gets an English SERP ($0.02) before a line of code; two or three tools per
+session is the honest rate; once the family exists, §7's sibling rule multiplies it by every
+verified market. kunavo, 2026-09-03: two tool URLs, 43 hand-written locale guides, six pages a
+round — the page-type mix set that ceiling, not the process.
+
 Judgment calls that matter:
 
 - **Three words are the whole quality bar: free, instant, no signup.** Tool above the fold,
@@ -893,7 +909,7 @@ and on a site with history it hands you free seeds (`gsc.topQueries` / `newQueri
 | a `rank.movers` row whose impressions collapsed with a flat position, or a `gsc.pages.movers` row with `status` ≠ 200 | **technical round, no keyword work**: restore or redirect the URL, request indexing, re-read next window. Shape B (top of this file) — links and new pages do not fix a dead URL |
 | `gsc.pages.unseen` rows whose `index` is not indexed | **indexing round**: internal links into them, sitemap `lastmod`, indexing requests; do not judge their keywords until a window after they are indexed. This row is also the batch gate (§7.3 gate 4): last batch mostly indexed → ship the next family; mostly `Crawled - currently not indexed` → no new pages this round, and retire what Google has refused twice |
 | `rank.pending` non-empty | those words are not measured yet — no branch applies to them this round; do not read their zero as "missed" |
-| `authority_limited` | placement only **on that theme** — another page on it will not move it (§6). The label is per keyword, not per site: production continues on every other theme. If every funded theme reads this way, the site is Shape C (top of this file): re-derive the ledger toward tool, long-tail and locale rows before pitching |
+| `authority_limited` | placement only **on that theme** — another page on it will not move it (§6). The label is per keyword, not per site: production continues on every other theme. If every funded theme reads this way, the site is Shape C (top of this file): re-derive the ledger toward tool, long-tail and locale rows before pitching — and a site with no tools builds them first (§7.2) |
 | `gsc.newQueries` non-empty | mini-discovery: fold them into the matrix, price the new cells, extend the tracked list if they survive the SERP check |
 | no movers two rounds running | the plan is stale, not slow — back to §1 and re-derive toward new page types; more waiting will not convert a flat line. The placement queue does not gate this: a queue that has waited on a human for two rounds is recorded as `blocked:` and handed over, and production goes on without it — placement is human-gated, building is not, and one must never hold the other (kunavo's four ready pitches held production for seven rounds) |