open/claude-design
1
0
Fork 0
Code Issues Releases 1 Activity

release(claude-design): v0.2.0 — Phase 9 critique-iterate

Browse source 
Version bump plus the docs sync that ships Phase 9.

- .claude-plugin/plugin.json: 0.1.0 -> 0.2.0
- CHANGELOG.md: [0.2.0] entry covering the shared 05-critique-iterate.md
  reference, the eight per-preset hint blocks, export-to-HTML as the input,
  operator-confirmed layer mapping, untrusted-paste rule, operator-attested
  exit, design:critique routing, accessibility deferred to v0.4, the new
  SC4 test and the dual-format SC1 dogfood check.
- README.md: version badge 0.1.0 -> 0.2.0, references badge 13 -> 14,
  trigger count 12 -> 16, a Non-goals bullet routing critique to Phase 9,
  and a v0.2.0 release-history entry.
- CLAUDE.md: Status v0.2.0, nine-phase surface line, Architecture heading,
  and a Scope-fence paragraph stating Phase 9 routes to design:critique
  without duplicating it.

Two corrections of stale claims found while sweeping:
- CLAUDE.md said the hint-block heading letter is (e)/(f); measured across
  all eight presets it is (f) x4 and (g) x4. The letter-free substring
  'Critique-iterate hint' remains the canonical anchor.
- The README badge said 13 references and 12 triggers; ground truth is 14
  and 16.

Historical records left verbatim: the [0.1.0] and [0.1.0-pre] CHANGELOG
entries, the v0.1.0 README release note, and docs/review-2026-06-20.md all
describe what v0.1.0 actually shipped.

Deviation from the plan's pinned check: the plan expects a literal
'## v0.2.0' CHANGELOG heading, but this file has used Keep a Changelog
bracketed headings since v0.1.0. The entry is '## [0.2.0]' to match the
file; no shipped test asserts either form.

No tag, no catalog bump. /trekreview is the release gate and comes next.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PX9V8V9qH5kG4XKVzCdchk
This commit is contained in:
Kjell Tore Guttormsen 2026-08-20 23:05:53 +02:00
commit f596c832c4
4 changed files with 38 additions and 12 deletions
Download patch file Download diff file Expand all files Collapse all files

11
README.md
View file

@ -58,10 +58,10 @@ Claude Code with plugin support, and access to `claude.ai/design`. No npm packag
*AI-generated: all content produced by Claude Code through dialog-driven development. [Full disclosure →](GOVERNANCE.md)*
![Version](https://img.shields.io/badge/version-0.1.0-blue)
![Version](https://img.shields.io/badge/version-0.2.0-blue)
![Platform](https://img.shields.io/badge/platform-Claude_Code_Plugin-purple)
![Skill](https://img.shields.io/badge/skills-1-green)
![References](https://img.shields.io/badge/references-13-green)
![References](https://img.shields.io/badge/references-14-green)
![License](https://img.shields.io/badge/license-MIT-lightgrey)
---
@ -101,6 +101,7 @@ By design, this plugin does not:
- **Store artifact history or version artifacts.** Claude Design has no version-tree primitive and this plugin does not invent one.
- **Cover adjacent Anthropic surfaces.** Classic Artifacts at `claude.ai`, Live Artifacts in Claude Cowork, custom visuals embedded in a chat reply are out of scope — see [skills/claude-design-facilitator/references/00-what-claude-design-is-and-isnt.md](skills/claude-design-facilitator/references/00-what-claude-design-is-and-isnt.md) for the disambiguation reference.
- **Duplicate Anthropic's `knowledge-work-plugins/design` plugin.** No `/critique`, no `/accessibility`, no `/ux-copy`, no `/research-synthesis`, no `/design-system`, no `/handoff`. The post-design lane belongs to Anthropic's plugin.
- **Critique the artifact itself.** Information-architecture, structural and copy problems that surface tweaks cannot fix are not diagnosed here. Phase 9 (critique-iterate) coaches the HTML export, hands the operator `design:critique @<file>`, and turns what comes back into a re-prompt aimed at the one prompt layer responsible. The critique belongs to Anthropic's plugin; the re-prompt is this one's job.
`tests/validate-plugin.sh` enforces the forbidden-command-name list mechanically.
@ -206,9 +207,9 @@ The full output of the session is a single fenced markdown block (Phase 6) plus
| Skill | Triggers | Output |
|-------|----------|--------|
| `claude-design-facilitator` | 12 natural-language phrases (full list in `.triggers.txt`); also explicit `/claude-design-facilitator` slash command | A copy-paste-ready Claude Design prompt block composed from the five-layer stack and the per-preset pattern, with follow-up-turn expectations |
| `claude-design-facilitator` | 16 natural-language phrases (full list in `.triggers.txt`); also explicit `/claude-design-facilitator` slash command | A copy-paste-ready Claude Design prompt block composed from the five-layer stack and the per-preset pattern, with follow-up-turn expectations |
No commands, no agents, no hooks, no MCP servers at v0.1. The single skill is the entire user-facing surface.
No commands, no agents, no hooks, no MCP servers at v0.2. The single skill is the entire user-facing surface.
---
@ -316,6 +317,8 @@ When a trigger fires, run `bash verify.sh --strict` after the update to confirm
Full release history: [CHANGELOG.md](CHANGELOG.md). The plugin follows [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
**v0.2.0 — 2026-08-20.** Phase 9 — critique-iterate. The facilitation flow is now nine phases: 1-8 to a shipped artifact, plus an optional post-ship loop that exports the artifact, routes it to `design:critique` in Anthropic's plugin, reads the returned critique as untrusted reference data, and drafts a re-prompt aimed at the single prompt layer responsible. One shared reference ([`05-critique-iterate.md`](skills/claude-design-facilitator/references/05-critique-iterate.md)) plus a thin `Critique-iterate hint` block in each of the eight per-preset files. Layer mapping is operator-confirmed judgment rather than a parser; accessibility findings are deferred to v0.4. New `tests/test-sc4-phase9-coverage.sh` (21 assertions), and the dogfood-log check now accepts either a v0.1 or a v0.2 Phase 9 block.
**v0.1.0 — 2026-05-17.** Initial public release. Single skill (`claude-design-facilitator`) with eight-phase facilitation flow, 12 natural-language trigger phrases, 13 reference files (5 foundation + 8 per-preset with evidence-grade labels), `.coverage.md` preset manifest plus Authoritative-claims registry, five verification scripts under `tests/` enforcing structural integrity / scope fence / skill description quality / per-preset coverage / Anthropic-domain citation discipline / operator dogfood log format, top-level `verify.sh` roll-up with `--strict` and `--quick` flags, MIT license, GOVERNANCE.md fork-and-own model.
The path from v0.1 to v1.0 is dogfood-driven. The v1.0 readiness criteria are multi-preset breadth, auto-fire validation in real natural-language requests, and two consecutive dogfood sessions with zero critical patches. The dogfood log itself is operator-local and never published.