design-playbook 0.9.2 → 0.10.0

package/README.md

@@ -9,7 +9,7 @@ Declarations + contracts — not a style CSV pack. Compose with [ui-ux-pro-max](
 Path of record (published) - the marketplace catalog lives at the **repo root**, not in this package:
 
 ```text
-/plugin marketplace add <owner>/<repo>
+/plugin marketplace add https://github.com/Bandersnatch0x/design-playbook.git
 /plugin install design-playbook@design-playbook
 ```
 
@@ -55,6 +55,7 @@ After install, skills and commands are **namespaced** by the plugin name:
 | `/design-playbook:design-io` | Full pipeline command |
 | `/design-playbook:ux-spec` | Spec-only command |
 | `/design-playbook:ui-review` | Review command |
+| `/design-playbook:run-review` | Cross-run review command |
 
 Bare `/design-io` is **not** the installed name — always use the `design-playbook:` prefix.
 
@@ -72,7 +73,7 @@ pi has no plugin namespace — skills are `/skill:<name>`, commands are bare `/<
 | --- | --- |
 | `/skill:design-playbook` | Orchestrator skill (model-invoked) |
 | `/skill:ux-spec` … `/skill:ui-evaluator` | Same eight skills as above |
-| `/design-io` · `/ux-spec` · `/ui-review` | Pipeline / spec-only / review commands |
+| `/design-io` · `/ux-spec` · `/ui-review` · `/run-review` | Pipeline / spec-only / review / cross-run commands |
 
 pi ships no built-in MCP, so `preview*` and `observe*` skip by default (ADR-0009 absent→skip; the pipeline still runs spec → picker → fill → craft → accept). To enable both gates, install an MCP adapter and register the bundled servers in your project `.mcp.json`:
 
@@ -116,7 +117,7 @@ pi install npm:pi-mcp-adapter
 .mcp.json              ← bundled MCP servers, launched via ${CLAUDE_PLUGIN_ROOT} (ADR-0009)
 mcp/{preview,evidence}/← MCP adapter runtimes (preview_prototype / execute_capture_plan)
 skills/<name>/SKILL.md ← model-invoked skills
-commands/<name>.md     ← slash commands (design-io, ux-spec, ui-review only)
+commands/<name>.md     ← slash commands (design-io, ux-spec, ui-review, run-review)
 codex/AGENTS.md        ← Codex bridge notes
 examples/              ← self-authored onboarding samples
 LICENSE · NOTICE       ← authored-only scope

package/commands/run-review.md

@@ -0,0 +1,30 @@
+---
+description: Cross-run review of Design I/O runs — gate table + repeat blockers
+---
+
+Cross-run **run review** over `.scratch/<run>/` dirs in the user project. Not a step of a single Design I/O run. Markdown only; report header **`run-review/v1`**.
+
+## Discover
+
+Scan user-side `.scratch/<run>/` (not monorepo `dogfood/*` globs). **Include** only dirs that have `point-back.md` (one table row each). List dirs without `point-back.md` as skipped + reason — they contribute **0** rows. If runs-with-point-back **< 2**, refuse and report that N.
+
+## Report (tables, in order)
+
+1. **Inclusion manifest** first: `path | status` (`included` / `skipped` + reason). Note: hash match ≠ honest transcription of `observed`.
+2. **Per-run table** — mandatory **run-path** column; other columns as needed; **gate** from real `validate_run.py` exit when the script is present (plugin install: `packages/design-playbook/scripts/validate_run.py` per run); else literal `not checked`. **Never** infer ok from "artifacts look complete".
+3. **Repeat blockers** — pure frequency table `count | runs | observed text` (verbatim first-seen text). A **repeat blocker** is the same normalized `observed` text recurring across runs (**counting, not judging**). Rows only where ledger `result != pass`. Grouping key = `observed` **casefold + whitespace-collapsed**, then **char-for-char** equality only; `count ≥ 2`. Literal differences stay separate; optional `similar:` pointer line, never merge counts. **`_none_` is normal** when nothing qualifies (do not loosen normalization to manufacture repeats).
+4. **Point-back** cites: path + verbatim `observed:` quote; no line numbers.
+5. Rollup numbers derived **row-by-row** from the tables above — no "overall it seems".
+
+Ledger row shape: **ui-evaluator** step 2 (do not restate).
+
+```
+禁止:
+- no new run ledger
+- no prose lessons / narrative "learning"
+- no auto-writeback to baseline
+- no semantic clustering of observed — verbatim grouping only
+```
+
+Scope:
+$ARGUMENTS

package/package.json

@@ -1,42 +1,42 @@
-{
-  "name": "design-playbook",
-  "version": "0.9.2",
-  "description": "Design I/O for coding agents: controllable UI generation via declarations (spec/domain/craft/design/components/template) and contracts (skill/evaluator). Use for product UI—console, dashboard, agent-ops, CJK-first apps.",
-  "keywords": [
-    "pi-package",
-    "ui",
-    "ux",
-    "design",
-    "design-io",
-    "design-playbook",
-    "spec",
-    "craft",
-    "evaluator",
-    "console",
-    "cjk"
-  ],
-  "license": "MIT",
-  "author": "Bandersnatch0x (https://github.com/Bandersnatch0x)",
-  "homepage": "https://github.com/Bandersnatch0x/design-playbook",
-  "repository": {
-    "type": "git",
-    "url": "git+https://github.com/Bandersnatch0x/design-playbook.git",
-    "directory": "packages/design-playbook"
-  },
-  "files": [
-    "skills",
-    "commands",
-    "mcp",
-    "NOTICE",
-    "!**/__pycache__"
-  ],
-  "pi": {
-    "skills": [
-      "./skills"
-    ],
-    "prompts": [
-      "./commands"
-    ],
-    "image": "https://raw.githubusercontent.com/Bandersnatch0x/design-playbook/main/packages/design-playbook/showcase/screenshots/hero.png"
-  }
-}
+{
+  "name": "design-playbook",
+  "version": "0.10.0",
+  "description": "Design I/O for coding agents: controllable UI generation via declarations (spec/domain/craft/design/components/template) and contracts (skill/evaluator). Use for product UI—console, dashboard, agent-ops, CJK-first apps.",
+  "keywords": [
+    "pi-package",
+    "ui",
+    "ux",
+    "design",
+    "design-io",
+    "design-playbook",
+    "spec",
+    "craft",
+    "evaluator",
+    "console",
+    "cjk"
+  ],
+  "license": "MIT",
+  "author": "Bandersnatch0x (https://github.com/Bandersnatch0x)",
+  "homepage": "https://github.com/Bandersnatch0x/design-playbook",
+  "repository": {
+    "type": "git",
+    "url": "git+https://github.com/Bandersnatch0x/design-playbook.git",
+    "directory": "packages/design-playbook"
+  },
+  "files": [
+    "skills",
+    "commands",
+    "mcp",
+    "NOTICE",
+    "!**/__pycache__"
+  ],
+  "pi": {
+    "skills": [
+      "./skills"
+    ],
+    "prompts": [
+      "./commands"
+    ],
+    "image": "https://raw.githubusercontent.com/Bandersnatch0x/design-playbook/main/packages/design-playbook/showcase/screenshots/hero.png"
+  }
+}

package/skills/design-playbook/SKILL.md

@@ -195,6 +195,8 @@ Invoke **ui-evaluator**. Issues must **point back** to a declaration.
 
 **Done when:** the report includes the criterion-shaped evidence ledger (`criterion / required / observed / result`) and findings as `issue / source / fix / severity`; the authoritative verdict completion criterion in `ui-evaluator` is met; **and** you show the user a short **run artifact index** (paths under `.scratch/<run>/`) so declaration products are discoverable — at minimum: `design-baseline/` (if triggered), `reference/` (if any), `spec.md`, `plan.md`, `decision-report.md`, `preview/` (if any), Fill surface path, `evidence/` (if any), `point-back.md`. One block is enough; do not only leave paths buried in tool logs.
 
+Cross-run review of multiple `.scratch/<run>/` runs lives in command **run-review** (cross-run, not a step of this run).
+
 Machine seam (optional local check): `python scripts/validate_run.py <spec.md> <point-back.md> [--preview-dir <preview/>] [--decision-report <report>] [--evidence-dir <evidence/>] [--run-root <run>]`.
 
 ## Recirculate