@jairodri/huckleberry-cli 1.0.0
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/agents/cos.json +20 -0
- package/agents/hermes/AGENTS.md +62 -0
- package/agents/hermes/SOUL.md +29 -0
- package/agents/hermes/fix-kimaki/SKILL.md +100 -0
- package/agents/hermes/hermes-fixer.md +115 -0
- package/agents/hermes.json +26 -0
- package/bin/huckleberry.js +4 -0
- package/kanban/SKILL.md +167 -0
- package/kanban/kanban +3 -0
- package/kanban/package.json +9 -0
- package/kanban/src/cli.mjs +553 -0
- package/kanban/src/dispatcher.mjs +479 -0
- package/kanban/src/server.mjs +108 -0
- package/kanban/templates/b2b-outreach.yaml +84 -0
- package/kanban/templates/blog-social-campaign.yaml +1161 -0
- package/kanban/templates/client-onboard.yaml +107 -0
- package/kanban/templates/cro-pipeline.yaml +296 -0
- package/kanban/templates/social-campaign.yaml +356 -0
- package/mcp/gemini-codegen-server.mjs +134 -0
- package/package.json +46 -0
- package/plugins/env-inject.ts +61 -0
- package/plugins/inkstone-gate.ts +55 -0
- package/scripts/heartbeat.sh +41 -0
- package/scripts/install-services.ps1 +82 -0
- package/scripts/install-services.sh +74 -0
- package/scripts/keep-awake.ps1 +48 -0
- package/scripts/keep-awake.sh +31 -0
- package/scripts/keepalive.sh +43 -0
- package/skills/curated.txt +68 -0
- package/src/cli.js +227 -0
- package/src/interview.js +78 -0
- package/src/postinstall.js +22 -0
- package/taskmaster/SKILL.md +78 -0
- package/taskmaster/dispatcher.mjs +146 -0
- package/taskmaster/next-task.js +123 -0
- package/taskmaster/update-status.js +118 -0
- package/templates/AGENTS.md +90 -0
- package/templates/COS_PROTOCOLS.md +71 -0
- package/templates/EXECUTIVE-MODE.md +29 -0
- package/templates/HEARTBEAT.md +42 -0
- package/templates/IDENTITY.md +22 -0
- package/templates/MEMORY.md +17 -0
- package/templates/ORG.md +33 -0
- package/templates/SOUL.md +43 -0
- package/templates/TOOLS.md +66 -0
- package/templates/business-context-gate.md +9 -0
- package/templates/evergreen.md +29 -0
- package/templates/opencode.json +80 -0
- package/wiki/discord.md +52 -0
- package/wiki/inkstone.md +64 -0
- package/wiki/kimaki.md +48 -0
- package/wiki/opencode.md +43 -0
|
@@ -0,0 +1,107 @@
|
|
|
1
|
+
name: Client Onboard (Entry Point)
|
|
2
|
+
description: >
|
|
3
|
+
Thin entry-point wrapper. Takes a website URL or business idea and stamps the full
|
|
4
|
+
CMO Pipeline to build a complete web presence from zero.
|
|
5
|
+
This is the "one command" clients use to start: give it a URL or idea, it runs everything.
|
|
6
|
+
Stamps: cmo-pipeline (orchestrator) → which stamps blog-social-campaign per content asset.
|
|
7
|
+
|
|
8
|
+
variables:
|
|
9
|
+
- name: website_url
|
|
10
|
+
required: false
|
|
11
|
+
description: Existing website URL to scrape for voice, visuals, and content structure
|
|
12
|
+
- name: business_idea
|
|
13
|
+
required: false
|
|
14
|
+
description: Business description if no website exists yet (e.g. "A solar charger brand for van lifers")
|
|
15
|
+
- name: product
|
|
16
|
+
required: true
|
|
17
|
+
description: Product or service name
|
|
18
|
+
- name: tenant
|
|
19
|
+
required: true
|
|
20
|
+
description: Brand/project tenant name (maps to a kimaki project)
|
|
21
|
+
- name: vertical
|
|
22
|
+
required: true
|
|
23
|
+
description: Industry vertical (e.g. "saas", "resorts", "e-commerce", "consumer goods")
|
|
24
|
+
- name: spoke_count
|
|
25
|
+
default: "10"
|
|
26
|
+
description: Target number of TOFU spoke pages
|
|
27
|
+
- name: cornerstone_count
|
|
28
|
+
default: "3"
|
|
29
|
+
description: Number of MOFU cornerstone articles
|
|
30
|
+
- name: platforms
|
|
31
|
+
default: twitter,linkedin,instagram,reddit
|
|
32
|
+
description: Comma-separated social platforms
|
|
33
|
+
- name: audience_hint
|
|
34
|
+
required: false
|
|
35
|
+
description: Quick audience description (e.g. "campers who overpay for gear")
|
|
36
|
+
- name: competitor_hint
|
|
37
|
+
required: false
|
|
38
|
+
description: Known competitors (comma-separated)
|
|
39
|
+
- name: launch_date
|
|
40
|
+
required: false
|
|
41
|
+
description: Target launch date (YYYY-MM-DD)
|
|
42
|
+
|
|
43
|
+
tasks:
|
|
44
|
+
- id: validate_input
|
|
45
|
+
title: "Validate onboarding input: {{product}}"
|
|
46
|
+
assignee: researcher
|
|
47
|
+
priority: 5
|
|
48
|
+
body: |
|
|
49
|
+
Validate the onboarding input for "{{product}}".
|
|
50
|
+
|
|
51
|
+
Check that at least ONE of the following is provided:
|
|
52
|
+
- website_url: "{{website_url}}"
|
|
53
|
+
- business_idea: "{{business_idea}}"
|
|
54
|
+
|
|
55
|
+
If neither is provided, BLOCK the task and ask the user:
|
|
56
|
+
"I need either a website URL to analyze, or a description of your business idea to get started."
|
|
57
|
+
|
|
58
|
+
If website_url is provided:
|
|
59
|
+
- Verify the URL is reachable
|
|
60
|
+
- Flag whether the site uses WordPress (affects publishing approach)
|
|
61
|
+
- Note: voice and visual extraction will happen in cmo-pipeline intake phase
|
|
62
|
+
|
|
63
|
+
If business_idea is provided:
|
|
64
|
+
- Note that foundation will be built from scratch (no existing brand to extract from)
|
|
65
|
+
- Flag that audience_hint and competitor_hint become more important
|
|
66
|
+
|
|
67
|
+
Produce a brief onboarding summary and pass to the stamp task.
|
|
68
|
+
|
|
69
|
+
- id: stamp_pipeline
|
|
70
|
+
title: "Launch CMO Pipeline: {{product}}"
|
|
71
|
+
assignee: launcher
|
|
72
|
+
parents: [validate_input]
|
|
73
|
+
priority: 5
|
|
74
|
+
body: |
|
|
75
|
+
Stamp the full CMO Pipeline for "{{product}}".
|
|
76
|
+
|
|
77
|
+
```bash
|
|
78
|
+
bun /Users/clawdbot/clawd/workspace/kimaki-kanban/src/cli.mjs stamp cmo-pipeline \
|
|
79
|
+
--var product="{{product}}" \
|
|
80
|
+
--var tenant="{{tenant}}" \
|
|
81
|
+
--var vertical="{{vertical}}" \
|
|
82
|
+
--var spoke_count="{{spoke_count}}" \
|
|
83
|
+
--var cornerstone_count="{{cornerstone_count}}" \
|
|
84
|
+
--var platforms="{{platforms}}" \
|
|
85
|
+
--var audience_hint="{{audience_hint}}" \
|
|
86
|
+
--var competitor_hint="{{competitor_hint}}" \
|
|
87
|
+
--var launch_date="{{launch_date}}" \
|
|
88
|
+
--var website_url="{{website_url}}" \
|
|
89
|
+
--var business_idea="{{business_idea}}" \
|
|
90
|
+
--db /Users/clawdbot/.kimaki/kanban.db
|
|
91
|
+
```
|
|
92
|
+
|
|
93
|
+
The CMO Pipeline will:
|
|
94
|
+
1. Phase 0: Intake (scrape site or process business idea)
|
|
95
|
+
2. Phase 1: Foundation (voice, audience, competitors, landscape, positioning, funnel map)
|
|
96
|
+
3. Phase 2: Visual identity
|
|
97
|
+
4. Phase 3: Strategy (keywords, silo structure, link-map.json)
|
|
98
|
+
5. Phase 4: Content production (stamps blog-social-campaign for each spoke + cornerstone)
|
|
99
|
+
6. Phase 5: Link network
|
|
100
|
+
7. Phase 6: Distribution + launch
|
|
101
|
+
8. Phase 7: Evergreen audit + learnings
|
|
102
|
+
|
|
103
|
+
Each content asset passes the full 10-layer Content Quality Standard.
|
|
104
|
+
See: workspace/kimaki-kanban/content-quality-standard.md
|
|
105
|
+
|
|
106
|
+
Report: the stamped pipeline's board ID and task list.
|
|
107
|
+
The dispatcher will pick up tasks automatically.
|
|
@@ -0,0 +1,296 @@
|
|
|
1
|
+
name: CRO Pipeline
|
|
2
|
+
description: >
|
|
3
|
+
Full conversion rate optimization pipeline from discovery to test plan.
|
|
4
|
+
Reads brand context from the evergreen system. If brand files are missing,
|
|
5
|
+
the discovery task asks the user targeted questions before proceeding.
|
|
6
|
+
Brand-agnostic — works for any site with any conversion goal.
|
|
7
|
+
|
|
8
|
+
variables:
|
|
9
|
+
- name: tenant
|
|
10
|
+
required: true
|
|
11
|
+
description: Brand/project tenant name (maps to a kimaki project)
|
|
12
|
+
- name: target_url
|
|
13
|
+
required: true
|
|
14
|
+
description: Primary URL or flow entry point to optimize
|
|
15
|
+
- name: conversion_goal
|
|
16
|
+
required: false
|
|
17
|
+
description: What counts as a conversion (signup, purchase, demo request, etc.)
|
|
18
|
+
- name: flow_type
|
|
19
|
+
default: auto
|
|
20
|
+
description: >
|
|
21
|
+
Flow type to optimize. Options: page, signup-flow, onboarding,
|
|
22
|
+
checkout, upgrade, full-funnel, auto (detect automatically)
|
|
23
|
+
- name: current_rate
|
|
24
|
+
required: false
|
|
25
|
+
description: Current conversion rate if known (e.g. "2.3%")
|
|
26
|
+
- name: traffic_source
|
|
27
|
+
default: mixed
|
|
28
|
+
description: Primary traffic source (organic, paid, email, social, mixed)
|
|
29
|
+
|
|
30
|
+
tasks:
|
|
31
|
+
- id: discovery
|
|
32
|
+
title: "CRO Discovery: {{target_url}}"
|
|
33
|
+
assignee: cro-strategist
|
|
34
|
+
skills: [page-cro, conversion-flow-cro]
|
|
35
|
+
priority: 3
|
|
36
|
+
body: |
|
|
37
|
+
## CRO Discovery for {{tenant}}
|
|
38
|
+
|
|
39
|
+
Target URL: {{target_url}}
|
|
40
|
+
Conversion Goal: {{conversion_goal}}
|
|
41
|
+
Flow Type: {{flow_type}}
|
|
42
|
+
Current Rate: {{current_rate}}
|
|
43
|
+
Traffic Source: {{traffic_source}}
|
|
44
|
+
|
|
45
|
+
## Step 1: Read Brand Context
|
|
46
|
+
|
|
47
|
+
Read these files from the tenant's brand directory BEFORE asking questions:
|
|
48
|
+
- `brand/audience.md` — Personas, pain points, motivation levels
|
|
49
|
+
- `brand/positioning.md` — Value props and differentiation
|
|
50
|
+
- `brand/voice-profile.md` — Brand voice for copy recommendations
|
|
51
|
+
- `brand/creative-kit.md` — Visual identity constraints
|
|
52
|
+
- `businesses/{tenant}/evergreen.md` — Business context, goals, revenue model
|
|
53
|
+
|
|
54
|
+
## Step 2: Identify Gaps
|
|
55
|
+
|
|
56
|
+
After reading brand files, note which are missing or incomplete.
|
|
57
|
+
If critical files are missing, BLOCK the task and ask the user:
|
|
58
|
+
- No audience.md → "Who is your ideal customer? What's their biggest pain point?"
|
|
59
|
+
- No positioning.md → "What makes you different from competitors? What's your main value proposition?"
|
|
60
|
+
- No voice-profile.md → "How would you describe your brand tone? (casual, professional, provocative, etc.)"
|
|
61
|
+
- No conversion goal → "What specific action do you want visitors to take?"
|
|
62
|
+
|
|
63
|
+
## Step 3: Map the Target
|
|
64
|
+
|
|
65
|
+
Visit {{target_url}} and map what exists:
|
|
66
|
+
- Page type (landing, homepage, pricing, signup, checkout)
|
|
67
|
+
- Current headline, CTA copy, form fields
|
|
68
|
+
- Trust signals present
|
|
69
|
+
- Friction points visible
|
|
70
|
+
- Mobile experience assessment
|
|
71
|
+
|
|
72
|
+
If flow_type is "auto", determine from the URL and page content whether this is:
|
|
73
|
+
- A single page optimization → page-cro
|
|
74
|
+
- A multi-step flow (signup/onboarding/checkout) → conversion-flow-cro
|
|
75
|
+
- A form-heavy page → form-cro
|
|
76
|
+
|
|
77
|
+
## Output
|
|
78
|
+
|
|
79
|
+
Write a discovery report covering:
|
|
80
|
+
1. Brand context summary (what was found, what was missing)
|
|
81
|
+
2. Current state assessment of the target URL/flow
|
|
82
|
+
3. Recommended CRO focus areas ranked by estimated impact
|
|
83
|
+
4. Which sub-audits to run (page, flow, form, onboarding, paywall)
|
|
84
|
+
|
|
85
|
+
- id: page_audit
|
|
86
|
+
title: "Page CRO Audit: {{target_url}}"
|
|
87
|
+
assignee: cro-auditor
|
|
88
|
+
skills: [page-cro]
|
|
89
|
+
parents: [discovery]
|
|
90
|
+
priority: 2
|
|
91
|
+
body: |
|
|
92
|
+
## Page CRO Audit
|
|
93
|
+
|
|
94
|
+
Target: {{target_url}}
|
|
95
|
+
Tenant: {{tenant}}
|
|
96
|
+
Conversion Goal: {{conversion_goal}}
|
|
97
|
+
Traffic Source: {{traffic_source}}
|
|
98
|
+
|
|
99
|
+
Use the discovery report from the parent task as context.
|
|
100
|
+
Read brand files (audience.md, positioning.md, voice-profile.md) for copy recommendations.
|
|
101
|
+
|
|
102
|
+
Audit the page across these dimensions (score each 1-10):
|
|
103
|
+
1. Value Proposition Clarity
|
|
104
|
+
2. Headline Effectiveness
|
|
105
|
+
3. CTA Placement, Copy, and Hierarchy
|
|
106
|
+
4. Visual Hierarchy and Scannability
|
|
107
|
+
5. Trust Signals and Social Proof
|
|
108
|
+
6. Objection Handling
|
|
109
|
+
7. Friction Points
|
|
110
|
+
|
|
111
|
+
For each dimension provide:
|
|
112
|
+
- Current state (what exists now)
|
|
113
|
+
- Issues found
|
|
114
|
+
- Specific fix recommendation
|
|
115
|
+
- 2-3 copy alternatives where applicable (matching brand voice)
|
|
116
|
+
|
|
117
|
+
Output: Scored audit with prioritized quick wins and test ideas.
|
|
118
|
+
|
|
119
|
+
- id: flow_audit
|
|
120
|
+
title: "Flow CRO Audit: {{target_url}}"
|
|
121
|
+
assignee: cro-auditor
|
|
122
|
+
skills: [conversion-flow-cro, signup-flow-cro, onboarding-cro]
|
|
123
|
+
parents: [discovery]
|
|
124
|
+
priority: 2
|
|
125
|
+
body: |
|
|
126
|
+
## Multi-Step Flow CRO Audit
|
|
127
|
+
|
|
128
|
+
Target: {{target_url}}
|
|
129
|
+
Tenant: {{tenant}}
|
|
130
|
+
Flow Type: {{flow_type}}
|
|
131
|
+
Conversion Goal: {{conversion_goal}}
|
|
132
|
+
|
|
133
|
+
Use the discovery report from the parent task as context.
|
|
134
|
+
Read brand files for voice and positioning guidance.
|
|
135
|
+
Read `references/flow-benchmarks.md` from the conversion-flow-cro skill for benchmark data.
|
|
136
|
+
|
|
137
|
+
Map every step in the conversion flow:
|
|
138
|
+
- Step name, URL/route, fields required, CTA copy
|
|
139
|
+
- Friction score per step (1-5)
|
|
140
|
+
- Drop-off risk assessment
|
|
141
|
+
|
|
142
|
+
For each step:
|
|
143
|
+
- Identify the #1 friction point
|
|
144
|
+
- Recommend a specific copy or UX fix
|
|
145
|
+
- Suggest where to add progress indicators, social proof, or motivation boosters
|
|
146
|
+
|
|
147
|
+
Apply progressive disclosure, smart defaults, and single-field starts where possible.
|
|
148
|
+
|
|
149
|
+
Output: Flow map with per-step drop-off risk, redesign recommendations, and benchmark comparison.
|
|
150
|
+
|
|
151
|
+
- id: form_audit
|
|
152
|
+
title: "Form CRO Audit: {{target_url}}"
|
|
153
|
+
assignee: cro-auditor
|
|
154
|
+
skills: [form-cro, signup-flow-cro]
|
|
155
|
+
parents: [discovery]
|
|
156
|
+
priority: 2
|
|
157
|
+
body: |
|
|
158
|
+
## Form CRO Audit
|
|
159
|
+
|
|
160
|
+
Target: {{target_url}}
|
|
161
|
+
Tenant: {{tenant}}
|
|
162
|
+
Conversion Goal: {{conversion_goal}}
|
|
163
|
+
|
|
164
|
+
Use the discovery report from the parent task as context.
|
|
165
|
+
Read brand files for voice guidance on label and placeholder copy.
|
|
166
|
+
|
|
167
|
+
Audit all forms on the target page/flow:
|
|
168
|
+
1. Field count vs. minimum required
|
|
169
|
+
2. Field order optimization
|
|
170
|
+
3. Label, placeholder, and microcopy quality
|
|
171
|
+
4. Error handling and validation UX
|
|
172
|
+
5. Mobile form experience
|
|
173
|
+
6. Social auth options
|
|
174
|
+
7. Trust elements near forms
|
|
175
|
+
|
|
176
|
+
For each form issue:
|
|
177
|
+
- Rate severity (high/medium/low)
|
|
178
|
+
- Provide specific fix
|
|
179
|
+
- Estimate impact on completion rate
|
|
180
|
+
|
|
181
|
+
Output: Form audit with field-by-field recommendations and redesigned form spec.
|
|
182
|
+
|
|
183
|
+
- id: redesign
|
|
184
|
+
title: "CRO Redesign Package: {{target_url}}"
|
|
185
|
+
assignee: cro-designer
|
|
186
|
+
skills: [copywriting]
|
|
187
|
+
parents: [page_audit, flow_audit, form_audit]
|
|
188
|
+
priority: 1
|
|
189
|
+
body: |
|
|
190
|
+
## CRO Redesign Package
|
|
191
|
+
|
|
192
|
+
Target: {{target_url}}
|
|
193
|
+
Tenant: {{tenant}}
|
|
194
|
+
|
|
195
|
+
Synthesize findings from the three parent audits (page, flow, form).
|
|
196
|
+
Read brand files for voice and positioning — ALL copy must match brand voice.
|
|
197
|
+
|
|
198
|
+
Produce a complete redesign package:
|
|
199
|
+
|
|
200
|
+
### 1. Quick Wins (implement today)
|
|
201
|
+
List changes that require no design/dev resources or minimal effort.
|
|
202
|
+
For each: what to change, what to change it to, expected impact.
|
|
203
|
+
|
|
204
|
+
### 2. Copy Redesign
|
|
205
|
+
- New headline options (3 variants with rationale)
|
|
206
|
+
- New CTA copy (3 variants)
|
|
207
|
+
- Supporting copy for each step/screen
|
|
208
|
+
- Microcopy for forms, errors, and progress indicators
|
|
209
|
+
|
|
210
|
+
### 3. UX Recommendations
|
|
211
|
+
- Field reduction plan (which fields to remove, defer, or make optional)
|
|
212
|
+
- Flow restructuring (step order, consolidation)
|
|
213
|
+
- Social proof placement map
|
|
214
|
+
- Trust signal additions
|
|
215
|
+
|
|
216
|
+
### 4. Implementation Priority
|
|
217
|
+
Rank all changes by: impact × effort, grouped into:
|
|
218
|
+
- Week 1 (quick wins + high-impact low-effort)
|
|
219
|
+
- Week 2-3 (medium-effort improvements)
|
|
220
|
+
- Month 2 (larger structural changes)
|
|
221
|
+
|
|
222
|
+
Output: Complete redesign document ready for implementation.
|
|
223
|
+
|
|
224
|
+
- id: test_plan
|
|
225
|
+
title: "A/B Test Plan: {{target_url}}"
|
|
226
|
+
assignee: cro-strategist
|
|
227
|
+
skills: [ab-test-setup]
|
|
228
|
+
parents: [redesign]
|
|
229
|
+
priority: 1
|
|
230
|
+
body: |
|
|
231
|
+
## A/B Test Plan
|
|
232
|
+
|
|
233
|
+
Target: {{target_url}}
|
|
234
|
+
Tenant: {{tenant}}
|
|
235
|
+
Current Rate: {{current_rate}}
|
|
236
|
+
|
|
237
|
+
Using the redesign package from the parent task, create a prioritized A/B test plan.
|
|
238
|
+
|
|
239
|
+
For each test:
|
|
240
|
+
- Hypothesis statement
|
|
241
|
+
- Control vs. variant description
|
|
242
|
+
- Primary and secondary metrics
|
|
243
|
+
- Estimated sample size needed
|
|
244
|
+
- Minimum test duration
|
|
245
|
+
- Expected impact range
|
|
246
|
+
|
|
247
|
+
Prioritize tests by:
|
|
248
|
+
1. Highest expected impact on conversion rate
|
|
249
|
+
2. Lowest implementation effort
|
|
250
|
+
3. Fastest time to statistical significance
|
|
251
|
+
|
|
252
|
+
Group tests into phases:
|
|
253
|
+
- Phase 1: Hero/headline/CTA tests (highest impact, fastest results)
|
|
254
|
+
- Phase 2: Form and flow changes
|
|
255
|
+
- Phase 3: Trust signals and social proof
|
|
256
|
+
- Phase 4: Structural/layout changes
|
|
257
|
+
|
|
258
|
+
Output: Test plan with phase grouping, sample sizes, and success criteria.
|
|
259
|
+
|
|
260
|
+
- id: handoff
|
|
261
|
+
title: "CRO Handoff: {{target_url}}"
|
|
262
|
+
assignee: cro-strategist
|
|
263
|
+
skills: [page-cro]
|
|
264
|
+
parents: [redesign, test_plan]
|
|
265
|
+
priority: 0
|
|
266
|
+
body: |
|
|
267
|
+
## CRO Handoff Summary
|
|
268
|
+
|
|
269
|
+
Target: {{target_url}}
|
|
270
|
+
Tenant: {{tenant}}
|
|
271
|
+
|
|
272
|
+
Read the parent task results (redesign package + test plan).
|
|
273
|
+
Produce a single executive summary covering:
|
|
274
|
+
|
|
275
|
+
1. Current state assessment (from discovery)
|
|
276
|
+
2. Top 5 recommended changes ranked by impact
|
|
277
|
+
3. Implementation timeline (what to do this week, this month, next month)
|
|
278
|
+
4. A/B test roadmap summary
|
|
279
|
+
5. Expected overall conversion rate improvement range
|
|
280
|
+
6. Key risks and dependencies
|
|
281
|
+
|
|
282
|
+
BLOCK this task — present the summary to the user for approval before anyone implements.
|
|
283
|
+
|
|
284
|
+
# Tournament Mode (Quality Gate v2.0)
|
|
285
|
+
#
|
|
286
|
+
# For any task that produces high-impact short-form copy (headlines, subject lines,
|
|
287
|
+
# hooks, CTAs, social post first lines), run the copywriting tournament BEFORE
|
|
288
|
+
# standard quality-gate:
|
|
289
|
+
#
|
|
290
|
+
# 1. Generate 3-5 variants
|
|
291
|
+
# 2. Spawn 5 parallel judge subagents (skeptic, scanner, buyer, expert, outsider)
|
|
292
|
+
# Each loads persona from: ~/.claude/skills/quality-gate/judge-personas/{name}.md
|
|
293
|
+
# 3. Aggregate scores, pick winner
|
|
294
|
+
# 4. Winner enters standard quality-gate (single reviewer, pass/fail)
|
|
295
|
+
#
|
|
296
|
+
# See: Skill quality-gate for full tournament prompt templates and integration patterns.
|