@hegemonart/get-design-done 1.37.2 → 1.38.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/.claude-plugin/marketplace.json +2 -2
- package/.claude-plugin/plugin.json +1 -1
- package/CHANGELOG.md +23 -0
- package/README.md +4 -0
- package/agents/design-verifier.md +1 -1
- package/agents/experiment-result-ingester.md +61 -0
- package/agents/user-research-synthesizer.md +65 -0
- package/connections/connections.md +7 -1
- package/connections/growthbook.md +110 -0
- package/connections/hotjar.md +110 -0
- package/connections/launchdarkly.md +83 -0
- package/connections/maze.md +130 -0
- package/connections/statsig.md +83 -0
- package/connections/usertesting.md +99 -0
- package/package.json +1 -1
- package/reference/design-variants.md +56 -0
- package/reference/registry.json +7 -0
- package/scripts/lib/ds-arms/design-arms-store.cjs +119 -0
- package/skills/brief/SKILL.md +8 -0
- package/skills/connections/SKILL.md +4 -4
- package/skills/connections/connections-onboarding.md +58 -4
- package/skills/design/SKILL.md +2 -1
|
@@ -9,7 +9,7 @@ last_updated: 2026-05-18
|
|
|
9
9
|
|
|
10
10
|
Source: extracted from `skills/connections/SKILL.md` (Phase 28.5 rework — D-10 extract-then-link).
|
|
11
11
|
The skill's load-bearing routing + invocation-mode dispatch stays in `../skills/connections/SKILL.md`;
|
|
12
|
-
this file holds the
|
|
12
|
+
this file holds the 33 probe scripts, bucket categorization, per-connection setup screen,
|
|
13
13
|
auto-run eligibility matrix, value-prop one-liners, and STATE.md / config.json write contracts.
|
|
14
14
|
|
|
15
15
|
# Connections Onboarding Procedure
|
|
@@ -27,7 +27,7 @@ this file does NOT duplicate them; it points at them by name.
|
|
|
27
27
|
|
|
28
28
|
---
|
|
29
29
|
|
|
30
|
-
## Step 1 — Probe all
|
|
30
|
+
## Step 1 — Probe all 33 connections
|
|
31
31
|
|
|
32
32
|
Run every probe below in order. MCP probes call `ToolSearch` first (deferred tools fail silently without it). Write every result to `STATE.md <connections>` when done.
|
|
33
33
|
|
|
@@ -238,7 +238,49 @@ ToolSearch({ query: "builder", max_results: 5 })
|
|
|
238
238
|
→ Non-empty → builder-io: available
|
|
239
239
|
```
|
|
240
240
|
|
|
241
|
-
|
|
241
|
+
**launchdarkly:** (Outcome — experiment-source; read-only A/B)
|
|
242
|
+
```
|
|
243
|
+
ToolSearch({ query: "launchdarkly", max_results: 5 }) (else check the platform API-key env)
|
|
244
|
+
→ Empty / GDD_DISABLE_LAUNCHDARKLY=1 → launchdarkly: not_configured
|
|
245
|
+
→ Non-empty / key set → launchdarkly: available
|
|
246
|
+
```
|
|
247
|
+
|
|
248
|
+
**statsig:** (Outcome — experiment-source; read-only)
|
|
249
|
+
```
|
|
250
|
+
ToolSearch({ query: "statsig", max_results: 5 }) (else check the platform API-key env)
|
|
251
|
+
→ Empty / GDD_DISABLE_STATSIG=1 → statsig: not_configured
|
|
252
|
+
→ Non-empty / key set → statsig: available
|
|
253
|
+
```
|
|
254
|
+
|
|
255
|
+
**growthbook:** (Outcome — experiment-source; self-host/cloud)
|
|
256
|
+
```
|
|
257
|
+
ToolSearch({ query: "growthbook", max_results: 5 }) (else check the platform API-key env)
|
|
258
|
+
→ Empty / GDD_DISABLE_GROWTHBOOK=1 → growthbook: not_configured
|
|
259
|
+
→ Non-empty / key set → growthbook: available
|
|
260
|
+
```
|
|
261
|
+
|
|
262
|
+
**usertesting:** (Outcome — user-research; pseudonymize-first)
|
|
263
|
+
```
|
|
264
|
+
ToolSearch({ query: "usertesting", max_results: 5 }) (else check the platform API-key env)
|
|
265
|
+
→ Empty / GDD_DISABLE_USERTESTING=1 → usertesting: not_configured
|
|
266
|
+
→ Non-empty / key set → usertesting: available
|
|
267
|
+
```
|
|
268
|
+
|
|
269
|
+
**maze:** (Outcome — user-research; pseudonymize-first)
|
|
270
|
+
```
|
|
271
|
+
ToolSearch({ query: "maze", max_results: 5 }) (else check the platform API-key env)
|
|
272
|
+
→ Empty / GDD_DISABLE_MAZE=1 → maze: not_configured
|
|
273
|
+
→ Non-empty / key set → maze: available
|
|
274
|
+
```
|
|
275
|
+
|
|
276
|
+
**hotjar:** (Outcome — user-research; indexed insights only)
|
|
277
|
+
```
|
|
278
|
+
ToolSearch({ query: "hotjar", max_results: 5 }) (else check the platform API-key env)
|
|
279
|
+
→ Empty / GDD_DISABLE_HOTJAR=1 → hotjar: not_configured
|
|
280
|
+
→ Non-empty / key set → hotjar: available
|
|
281
|
+
```
|
|
282
|
+
|
|
283
|
+
After all 33 probes complete, merge results into `STATE.md <connections>`. Preserve the three-value schema verbatim (`available | unavailable | not_configured`). Do not add new values.
|
|
242
284
|
|
|
243
285
|
---
|
|
244
286
|
|
|
@@ -334,6 +376,12 @@ One-line value props (use verbatim):
|
|
|
334
376
|
| v0-dev | generator — Vercel v0 prompt→component (MCP-first → REST + V0_API_KEY) |
|
|
335
377
|
| plasmic | generator + canvas — emit code + read the Plasmic canvas |
|
|
336
378
|
| builder-io | generator — Builder.io Visual Copilot (pull-only this phase) |
|
|
379
|
+
| launchdarkly | outcome — read A/B results → design_arms posterior (experiment-source) |
|
|
380
|
+
| statsig | outcome — read experiment/pulse results → design_arms (experiment-source) |
|
|
381
|
+
| growthbook | outcome — read experiment results → design_arms (experiment-source; OSS) |
|
|
382
|
+
| usertesting | outcome — read test reports → brief findings (user-research; pseudonymized) |
|
|
383
|
+
| maze | outcome — read usability metrics → brief findings (user-research; pseudonymized) |
|
|
384
|
+
| hotjar | outcome — read indexed insights → brief findings (user-research; pseudonymized) |
|
|
337
385
|
|
|
338
386
|
---
|
|
339
387
|
|
|
@@ -361,7 +409,7 @@ options:
|
|
|
361
409
|
- "Exit" → emit ## CONNECTIONS COMPLETE, exit
|
|
362
410
|
```
|
|
363
411
|
|
|
364
|
-
If recommended bucket is empty, swap that option for "Show all
|
|
412
|
+
If recommended bucket is empty, swap that option for "Show all 33 and pick."
|
|
365
413
|
|
|
366
414
|
---
|
|
367
415
|
|
|
@@ -433,6 +481,12 @@ options:
|
|
|
433
481
|
| v0-dev | `claude mcp add v0 ...` (or V0_API_KEY) | ✓ yes | Reversible; MCP-first → REST fallback |
|
|
434
482
|
| plasmic | `claude mcp add plasmic ...` (or token) | ✓ yes | Reversible; dual canvas+generator |
|
|
435
483
|
| builder-io | `claude mcp add builder ...` (or BUILDER_API_KEY) | ✓ yes | Reversible; pull-only this phase |
|
|
484
|
+
| launchdarkly | (set `LAUNCHDARKLY_API_KEY` / add the LaunchDarkly MCP) | ✓ yes | Reversible; read-only experiment-source |
|
|
485
|
+
| statsig | (set `STATSIG_API_KEY` / add the Statsig MCP) | ✓ yes | Reversible; read-only experiment-source |
|
|
486
|
+
| growthbook | (set `GROWTHBOOK_API_KEY` [+ host] / add the MCP) | ✓ yes | Reversible; self-host or cloud |
|
|
487
|
+
| usertesting | (set `USERTESTING_API_KEY` / add the MCP) | ✓ yes | Reversible; read-only; pseudonymize-first |
|
|
488
|
+
| maze | (set `MAZE_API_KEY` / add the MCP) | ✓ yes | Reversible; read-only; pseudonymize-first |
|
|
489
|
+
| hotjar | (set `HOTJAR_API_KEY` / add the MCP) | ✓ yes | Reversible; indexed insights only; pseudonymize-first |
|
|
436
490
|
|
|
437
491
|
For non-auto-run connections, hide the "Run install command now" option entirely in 5.3.
|
|
438
492
|
|
package/skills/design/SKILL.md
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
---
|
|
2
2
|
name: design
|
|
3
3
|
description: "Stage 4 of 5 orchestrator that reads DESIGN-PLAN.md, partitions tasks by wave + parallel-safe flag, and spawns design-executor agents with the appropriate isolation (worktree for parallel batches, in-place for sequential tail). Use when DESIGN-PLAN.md is approved and ready for implementation."
|
|
4
|
-
argument-hint: "[--auto] [--parallel]"
|
|
4
|
+
argument-hint: "[--auto] [--parallel] [--variants N]"
|
|
5
5
|
user-invocable: true
|
|
6
6
|
tools: Read, Write, Bash, Grep, Glob, Task, AskUserQuestion, mcp__gdd_state__get, mcp__gdd_state__transition_stage, mcp__gdd_state__update_progress, mcp__gdd_state__set_status, mcp__gdd_state__add_blocker, mcp__gdd_state__resolve_blocker, mcp__gdd_state__checkpoint
|
|
7
7
|
---
|
|
@@ -28,6 +28,7 @@ Detail: `./design-procedure.md` §Stage entry.
|
|
|
28
28
|
|
|
29
29
|
- `--auto` -> `auto_mode=true` (no mid-stage prompts; architectural deviations stop the individual task but continue the rest).
|
|
30
30
|
- `--parallel` -> `parallel_mode=true` (use worktree isolation for `Parallel: true` tasks).
|
|
31
|
+
- `--variants N` -> `variants_mode` (default N=2): for tasks that build a user-facing surface, the executor emits **N competing, hypothesis-tagged variants** (`<variant id component pattern hypothesis>`) instead of one. Before generating, consult the `design_arms` posterior (`scripts/lib/ds-arms/design-arms-store.cjs`) to bias toward patterns that have won prior A/B / user-research outcomes — **advisory, never directive** (the user's explicit choice always wins). The tagged variants flow to A/B (LaunchDarkly/Statsig/GrowthBook) for outcome ingestion. Full schema + the outcome loop: `../../reference/design-variants.md`.
|
|
31
32
|
- **Directionally-open check** (skipped if `auto_mode`): scan DESIGN-PLAN.md for tasks whose criteria read "explore N directions" / "pick a visual approach" and suggest `/gdd:sketch` first.
|
|
32
33
|
- **Project-local conventions**: include any `./.claude/skills/design-*-conventions.md` and `~/.claude/gdd/global-skills/*.md` in every executor's `<required_reading>` — global conventions inform but do not override project-local D-XX decisions.
|
|
33
34
|
- **`.stories.tsx` stub**: after each new component file is created by the executor, emit a CSF stub alongside if `.storybook/` exists or `"storybook"` is in `package.json`, even with the dev server offline. Detail: `./design-procedure.md` §.stories.tsx Stub.
|