@clize/clize 0.30.2 → 0.30.3

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@clize/clize",
-  "version": "0.30.2",
+  "version": "0.30.3",
   "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

@@ -31,13 +31,13 @@ single most common failure shape, measured on our own tenant zero: nine weeks of
 at 120%, live referring domains at zero, every non-brand keyword flat at position 65+. If your
 plan is all pages and no placement, it is that plan.
 
-**§10 is the cycle that strings everything together, and every session starts there — with
-the free entry read (`clize seo check`), not with buying data.** §1–§3 are how to run a
-discovery round when the read calls for one, §4–§5 read the discovery data, §6–§7 turn rows
-into placement and pages, §8 is writing for AI engines, §9 reads the scoreboard. If you are
-here because someone asked for "keyword opportunities", still start at §10's entry read — it
-is free and may hand you seeds — then §1: the most common failure of this product is not
-misreading a number, it is testing a keyword set that was never enumerated.
+**§10 is the cycle that strings everything together. If a site exists, every session starts
+there — with the free entry read (`clize seo check`), not with buying data. No site yet (or
+nothing to point `--domain` at)? Then the cycle hasn't started: begin at §1, ship the first
+batch, and enter the cycle at your first enrolled check.** §1–§3 are how to run a discovery
+round, §4–§5 read the discovery data, §6–§7 turn rows into placement and pages, §8 is writing
+for AI engines, §9 reads the scoreboard. Either way, the most common failure of this product
+is not misreading a number, it is testing a keyword set that was never enumerated (§1).
 
 ---
 
@@ -587,9 +587,14 @@ double down where a return signal appeared, stop where none did, discover only w
 hands you something new. Measurement is the allocator, not a dashboard.
 
 ```
-clize seo check            # free — every session starts here
+clize seo check            # free — every session on an existing site starts here
 ```
 
+**Domain of the loop: a site that exists.** `check` needs `--domain` (or a clize project
+directory) — with neither it errors, correctly. No site yet, or a brand-new venture with
+nothing deployed? That is pre-cycle work: run discovery (§1) to decide what to build, ship
+the first batch (§7), enroll the keywords — the cycle starts at that first check.
+
 **Round one is not special — it is the round that reads zero.** The zeros are your baseline,
 and the entry read still pays before you spend a cent: it surfaces the GSC service-account
 setup (a 2-minute human step — hand it to your human now, so it runs in parallel with the
@@ -601,7 +606,7 @@ Branch on what the read says:
 | the read says | this round is |
 | --- | --- |
 | nothing yet (round one), or `demand_no_surface` across the board | **discovery + first bets**: `seo competitors <2–3 rivals>` → capability matrix (§1.2, seeded with anything the entry read gave you) → `seo keywords <the whole list>` in one call → drop `wall` → `seo serp` each survivor → worksheet (§3) → first batch ≤5 pages through the §7 pipeline, specs from `build site start`, prose per §8 → enroll: `seo check --keywords <survivors> --brand <you>` → start placement (§6) immediately — do not wait for rankings to "settle"; that wait is how sites end up at 120% on-page and zero authority. Total ≈ $1 |
-| `movers` rising, or a page drawing clicks / traffic / a first AI referral | **double down** — the branch the whole flywheel exists for: enumerate that keyword's sibling cells and price only the new ones (old ones are cached), add internal links into the winning page, add placement positions on that theme. Returns compound where the signal already is |
+| `movers` rising, a `rank.latest` row's clicks first going non-zero, or `traffic.ai.total` first non-zero | **double down** — the branch the whole flywheel exists for: enumerate that keyword's sibling cells and price only the new ones (old ones are cached), add internal links into that keyword's page (the worksheet's `target page` column is the query→page map — check itself reports queries, not pages), add placement positions on that theme. Returns compound where the signal already is |
 | `pre_emergence` | write that page this round — demand is ahead of the tools |
 | `authority_limited` | placement only; skip discovery *and* production — another page on the theme will not move it (§6) |
 | `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 |