tribunal-kit 4.3.1 → 4.4.1
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/.agent/agents/api-architect.md +66 -66
- package/.agent/agents/db-latency-auditor.md +216 -216
- package/.agent/agents/precedence-reviewer.md +250 -250
- package/.agent/agents/resilience-reviewer.md +88 -88
- package/.agent/agents/schema-reviewer.md +67 -67
- package/.agent/agents/throughput-optimizer.md +299 -299
- package/.agent/agents/ui-ux-auditor.md +292 -292
- package/.agent/agents/vitals-reviewer.md +223 -223
- package/.agent/scripts/_colors.js +18 -18
- package/.agent/scripts/_utils.js +42 -42
- package/.agent/scripts/append_flow.js +72 -72
- package/.agent/scripts/auto_preview.js +197 -197
- package/.agent/scripts/bundle_analyzer.js +290 -290
- package/.agent/scripts/case_law_manager.js +17 -6
- package/.agent/scripts/checklist.js +266 -266
- package/.agent/scripts/colors.js +17 -17
- package/.agent/scripts/compress_skills.js +141 -141
- package/.agent/scripts/consolidate_skills.js +149 -149
- package/.agent/scripts/context_broker.js +611 -609
- package/.agent/scripts/deep_compress.js +150 -150
- package/.agent/scripts/dependency_analyzer.js +272 -272
- package/.agent/scripts/graph_builder.js +151 -37
- package/.agent/scripts/graph_visualizer.js +384 -0
- package/.agent/scripts/inner_loop_validator.js +451 -465
- package/.agent/scripts/lint_runner.js +187 -187
- package/.agent/scripts/minify_context.js +100 -100
- package/.agent/scripts/mutation_runner.js +280 -0
- package/.agent/scripts/patch_skills_meta.js +156 -156
- package/.agent/scripts/patch_skills_output.js +244 -244
- package/.agent/scripts/schema_validator.js +297 -297
- package/.agent/scripts/security_scan.js +303 -303
- package/.agent/scripts/session_manager.js +276 -276
- package/.agent/scripts/skill_evolution.js +644 -644
- package/.agent/scripts/skill_integrator.js +313 -313
- package/.agent/scripts/strengthen_skills.js +193 -193
- package/.agent/scripts/strip_tribunal.js +47 -47
- package/.agent/scripts/swarm_dispatcher.js +360 -360
- package/.agent/scripts/test_runner.js +193 -193
- package/.agent/scripts/utils.js +32 -32
- package/.agent/scripts/verify_all.js +257 -256
- package/.agent/skills/app-builder/templates/astro-static/TEMPLATE.md +1 -1
- package/.agent/skills/app-builder/templates/chrome-extension/TEMPLATE.md +1 -1
- package/.agent/skills/app-builder/templates/cli-tool/TEMPLATE.md +1 -1
- package/.agent/skills/app-builder/templates/electron-desktop/TEMPLATE.md +1 -1
- package/.agent/skills/app-builder/templates/express-api/TEMPLATE.md +1 -1
- package/.agent/skills/app-builder/templates/flutter-app/TEMPLATE.md +1 -1
- package/.agent/skills/app-builder/templates/monorepo-turborepo/TEMPLATE.md +1 -1
- package/.agent/skills/app-builder/templates/nextjs-fullstack/TEMPLATE.md +1 -1
- package/.agent/skills/app-builder/templates/nextjs-saas/TEMPLATE.md +1 -1
- package/.agent/skills/app-builder/templates/nextjs-static/TEMPLATE.md +1 -1
- package/.agent/skills/app-builder/templates/nuxt-app/TEMPLATE.md +1 -1
- package/.agent/skills/app-builder/templates/python-fastapi/TEMPLATE.md +1 -1
- package/.agent/skills/app-builder/templates/react-native-app/TEMPLATE.md +1 -1
- package/.agent/skills/doc.md +1 -1
- package/.agent/skills/knowledge-graph/SKILL.md +32 -16
- package/.agent/skills/testing-patterns/SKILL.md +19 -2
- package/.agent/skills/ui-ux-pro-max/SKILL.md +480 -43
- package/.agent/workflows/generate.md +183 -183
- package/.agent/workflows/tribunal-speed.md +183 -183
- package/README.md +1 -1
- package/bin/tribunal-kit.js +134 -17
- package/package.json +6 -3
- package/scripts/changelog.js +167 -167
- package/scripts/sync-version.js +81 -81
- package/.agent/scripts/__pycache__/_colors.cpython-311.pyc +0 -0
- package/.agent/scripts/__pycache__/_utils.cpython-311.pyc +0 -0
- package/.agent/scripts/__pycache__/case_law_manager.cpython-311.pyc +0 -0
|
@@ -1,183 +1,183 @@
|
|
|
1
|
-
---
|
|
2
|
-
description: Generate code using the full Tribunal Anti-Hallucination pipeline. Maker generates grounded in real project context at low temperature → domain-selected reviewers audit in parallel → Human Gate for final approval. Nothing is written to disk without explicit approval.
|
|
3
|
-
---
|
|
4
|
-
|
|
5
|
-
# /generate — Hallucination-Free Code Generation
|
|
6
|
-
|
|
7
|
-
$ARGUMENTS
|
|
8
|
-
|
|
9
|
-
---
|
|
10
|
-
|
|
11
|
-
## When to Use /generate
|
|
12
|
-
|
|
13
|
-
| Use `/generate` when... | Use something else when... |
|
|
14
|
-
|:---|:---|
|
|
15
|
-
| New code needs to be written from scratch | Existing code needs modification → `/enhance` |
|
|
16
|
-
| A single focused piece of code is needed | Multi-domain build → `/create` or `/swarm` |
|
|
17
|
-
| A safe, reviewed snippet is required | You want to understand options first → `/plan` |
|
|
18
|
-
| You need a quick but Tribunal-reviewed piece | Full project structure needed → `/create` |
|
|
19
|
-
|
|
20
|
-
---
|
|
21
|
-
|
|
22
|
-
## Pipeline Flow
|
|
23
|
-
|
|
24
|
-
```
|
|
25
|
-
Your request
|
|
26
|
-
│
|
|
27
|
-
▼
|
|
28
|
-
[Phase 6] Context Broker — Skill Selection
|
|
29
|
-
├── Scores all 90+ skills against your task keywords
|
|
30
|
-
├── Level 0 (Essential): top skills — full content, injected first
|
|
31
|
-
├── Level 1 (Supplementary): medium relevance — key rules only
|
|
32
|
-
├── Level 2 (Available): listed for reference only
|
|
33
|
-
└── Large models: Essential + Supplementary | Small models: Essential only
|
|
34
|
-
│
|
|
35
|
-
▼
|
|
36
|
-
Context scan (MANDATORY before first line of code)
|
|
37
|
-
├── Read package.json → verify all imports exist
|
|
38
|
-
├── Read tsconfig.json → understand strictness, paths aliases
|
|
39
|
-
├── Read referenced files → understand actual data shapes
|
|
40
|
-
└── Read .env.example → know available environment variables
|
|
41
|
-
│
|
|
42
|
-
▼
|
|
43
|
-
Maker generates at temperature 0.1
|
|
44
|
-
├── Only methods verified in official docs
|
|
45
|
-
├── Only packages in package.json
|
|
46
|
-
├── // VERIFY: [reason] on any uncertain call
|
|
47
|
-
└── No full application generation — modules only
|
|
48
|
-
│
|
|
49
|
-
▼
|
|
50
|
-
[Phase 6] Inner-Loop Validator (AUTO — runs before you see the code)
|
|
51
|
-
├── Scans generated snippet for OWASP patterns (critical/high/medium/low)
|
|
52
|
-
├── Runs structural heuristics (empty catch, throw strings, env without fallback)
|
|
53
|
-
├── Verdict: APPROVED / WARNING / REJECTED
|
|
54
|
-
│ ├── APPROVED → continues to Tribunal Review
|
|
55
|
-
│ ├── WARNING → noted, continues with flag
|
|
56
|
-
│ └── REJECTED → Maker auto-corrects (up to 2 inner-loop attempts)
|
|
57
|
-
└── Only clean code reaches the Tribunal reviewers
|
|
58
|
-
│
|
|
59
|
-
▼
|
|
60
|
-
Tribunal Reviewers run in parallel (auto-selected by keyword)
|
|
61
|
-
│
|
|
62
|
-
▼
|
|
63
|
-
Human Gate — verdicts shown + unified diff
|
|
64
|
-
Y = write to disk | N = discard | R = revise with feedback
|
|
65
|
-
```
|
|
66
|
-
|
|
67
|
-
---
|
|
68
|
-
|
|
69
|
-
## What the Maker Is Not Allowed to Do
|
|
70
|
-
|
|
71
|
-
```
|
|
72
|
-
❌ Import a package not in package.json
|
|
73
|
-
❌ Call a method not verified in official documentation
|
|
74
|
-
❌ Use TypeScript 'any' without an explanation comment
|
|
75
|
-
❌ Generate an entire application in one shot
|
|
76
|
-
❌ Guess at database column or table names (read schema first)
|
|
77
|
-
❌ Fabricate API response shapes (read existing types first)
|
|
78
|
-
❌ Assume environment variables exist (read .env.example first)
|
|
79
|
-
❌ Use Next.js 14 patterns in a Next.js 15 project (check version!)
|
|
80
|
-
❌ Use React 18 hooks in a React 19 project (useFormState → useActionState)
|
|
81
|
-
❌ Use framer-motion v6 API in a v12 project (exitBeforeEnter → mode="wait")
|
|
82
|
-
❌ Use raw useEffect for GSAP — always useGSAP from @gsap/react
|
|
83
|
-
❌ Hallucinate LLM model names — verify against provider's current model list
|
|
84
|
-
```
|
|
85
|
-
|
|
86
|
-
When unsure: write `// VERIFY: [specific reason]` instead of hallucinating.
|
|
87
|
-
|
|
88
|
-
---
|
|
89
|
-
|
|
90
|
-
## Reviewer Auto-Selection
|
|
91
|
-
|
|
92
|
-
**Always active:**
|
|
93
|
-
```
|
|
94
|
-
precedence-reviewer→ Enforces repository Case Law and past rejections (Runs First)
|
|
95
|
-
logic-reviewer → Hallucinated methods, undefined refs, impossible logic
|
|
96
|
-
security-auditor → OWASP vulnerabilities, hardcoded secrets, injection
|
|
97
|
-
```
|
|
98
|
-
|
|
99
|
-
**Auto-activated by keywords:**
|
|
100
|
-
|
|
101
|
-
| Keyword in request | Additional Reviewers |
|
|
102
|
-
|:---|:---|
|
|
103
|
-
| `api`, `route`, `endpoint`, `handler`, `server action` | `dependency-reviewer` + `type-safety-reviewer` |
|
|
104
|
-
| `sql`, `query`, `database`, `prisma`, `drizzle`, `orm` | `sql-reviewer` |
|
|
105
|
-
| `component`, `hook`, `react`, `vue`, `jsx`, `tsx` | `frontend-reviewer` + `type-safety-reviewer` + `ui-ux-auditor` |
|
|
106
|
-
| `ui`, `design`, `landing`, `page`, `layout`, `style`, `css` | `ui-ux-auditor` + `accessibility-reviewer` |
|
|
107
|
-
| `animation`, `gsap`, `framer`, `motion`, `scroll` | `frontend-reviewer` + `performance-reviewer` + `ui-ux-auditor` |
|
|
108
|
-
| `test`, `spec`, `vitest`, `jest`, `playwright` | `test-coverage-reviewer` |
|
|
109
|
-
| `slow`, `optimize`, `cache`, `performance`, `bundle` | `performance-reviewer` |
|
|
110
|
-
| `mobile`, `react native`, `expo` | `mobile-reviewer` |
|
|
111
|
-
| `llm`, `openai`, `anthropic`, `gemini`, `embedding`, `ai` | `ai-code-reviewer` |
|
|
112
|
-
| `aria`, `wcag`, `a11y`, `accessibility` | `accessibility-reviewer` + `ui-ux-auditor` |
|
|
113
|
-
| `import`, `package`, `npm`, `require` | `dependency-reviewer` |
|
|
114
|
-
|
|
115
|
-
> For maximum safety on critical code: use `/tribunal-full` for all 11 reviewers simultaneously.
|
|
116
|
-
|
|
117
|
-
---
|
|
118
|
-
|
|
119
|
-
## Reviewer Verdicts
|
|
120
|
-
|
|
121
|
-
| Verdict | Meaning | What Happens |
|
|
122
|
-
|:---|:---|:---|
|
|
123
|
-
| `✅ APPROVED` | No issues found | Proceeds to Human Gate |
|
|
124
|
-
| `⚠️ WARNING` | Non-blocking issue | Human Gate shown with warning highlighted |
|
|
125
|
-
| `❌ REJECTED` | Blocking issue | Maker revises before Human Gate |
|
|
126
|
-
|
|
127
|
-
**Retry limit:** Maker is revised up to 3 times per REJECTED verdict. After 3 failures, the session halts and reports to the user with full failure history. No silent failures.
|
|
128
|
-
|
|
129
|
-
---
|
|
130
|
-
|
|
131
|
-
## Output Format
|
|
132
|
-
|
|
133
|
-
```
|
|
134
|
-
━━━ Tribunal: [Domain] ━━━━━━━━━━━━━━━━━━━━━━
|
|
135
|
-
|
|
136
|
-
Active reviewers: logic · security · [others]
|
|
137
|
-
|
|
138
|
-
[Generated code with // VERIFY: tags where uncertain]
|
|
139
|
-
|
|
140
|
-
━━━ Verdicts ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
|
|
141
|
-
|
|
142
|
-
logic-reviewer: ✅ APPROVED
|
|
143
|
-
security-auditor: ✅ APPROVED
|
|
144
|
-
dependency-reviewer: ⚠️ WARNING — lodash not in package.json
|
|
145
|
-
|
|
146
|
-
━━━ Warnings ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
|
|
147
|
-
|
|
148
|
-
dependency-reviewer:
|
|
149
|
-
⚠️ Medium — Line 3: 'lodash' imported but not in package.json
|
|
150
|
-
Fix: npm install lodash OR use built-in Array methods
|
|
151
|
-
|
|
152
|
-
━━━ Human Gate ━━━━━━━━━━━━━━━━━━━━━━━━━━━━
|
|
153
|
-
Write to disk? Y = approve | N = discard | R = revise with feedback
|
|
154
|
-
```
|
|
155
|
-
|
|
156
|
-
---
|
|
157
|
-
|
|
158
|
-
| After /generate shows... | Go to |
|
|
159
|
-
|:---|:---|
|
|
160
|
-
| Multiple files need changing | `/enhance` for impact-zone analysis |
|
|
161
|
-
| Security-critical code was generated | `/tribunal-full` for maximum coverage |
|
|
162
|
-
| DB queries were generated | `/tribunal-database` |
|
|
163
|
-
| New API routes were generated | `/tribunal-backend` |
|
|
164
|
-
| Animation/motion code generated | `/tribunal-frontend` |
|
|
165
|
-
| Tests need to be written next | `/test` |
|
|
166
|
-
| Something was rejected 3 times | Escalate to human with failure report |
|
|
167
|
-
|
|
168
|
-
---
|
|
169
|
-
|
|
170
|
-
## Usage Examples
|
|
171
|
-
|
|
172
|
-
```
|
|
173
|
-
/generate a JWT middleware for Express with HS256 algorithm enforcement
|
|
174
|
-
/generate a Prisma query for users with their published posts in last 30 days
|
|
175
|
-
/generate a debounced search hook in React 19 using useDeferredValue
|
|
176
|
-
/generate a parameterized SQL query for paginated order history
|
|
177
|
-
/generate a Zod schema for email + password + role login input
|
|
178
|
-
/generate a Server Action for creating a product with image upload
|
|
179
|
-
/generate a rate-limited fetch wrapper using @upstash/ratelimit
|
|
180
|
-
/generate a Framer Motion page transition with shared element (layoutId)
|
|
181
|
-
/generate a GSAP ScrollTrigger timeline with useGSAP React hook
|
|
182
|
-
/generate an OpenAI structured output call with Zod schema validation
|
|
183
|
-
```
|
|
1
|
+
---
|
|
2
|
+
description: Generate code using the full Tribunal Anti-Hallucination pipeline. Maker generates grounded in real project context at low temperature → domain-selected reviewers audit in parallel → Human Gate for final approval. Nothing is written to disk without explicit approval.
|
|
3
|
+
---
|
|
4
|
+
|
|
5
|
+
# /generate — Hallucination-Free Code Generation
|
|
6
|
+
|
|
7
|
+
$ARGUMENTS
|
|
8
|
+
|
|
9
|
+
---
|
|
10
|
+
|
|
11
|
+
## When to Use /generate
|
|
12
|
+
|
|
13
|
+
| Use `/generate` when... | Use something else when... |
|
|
14
|
+
|:---|:---|
|
|
15
|
+
| New code needs to be written from scratch | Existing code needs modification → `/enhance` |
|
|
16
|
+
| A single focused piece of code is needed | Multi-domain build → `/create` or `/swarm` |
|
|
17
|
+
| A safe, reviewed snippet is required | You want to understand options first → `/plan` |
|
|
18
|
+
| You need a quick but Tribunal-reviewed piece | Full project structure needed → `/create` |
|
|
19
|
+
|
|
20
|
+
---
|
|
21
|
+
|
|
22
|
+
## Pipeline Flow
|
|
23
|
+
|
|
24
|
+
```
|
|
25
|
+
Your request
|
|
26
|
+
│
|
|
27
|
+
▼
|
|
28
|
+
[Phase 6] Context Broker — Skill Selection
|
|
29
|
+
├── Scores all 90+ skills against your task keywords
|
|
30
|
+
├── Level 0 (Essential): top skills — full content, injected first
|
|
31
|
+
├── Level 1 (Supplementary): medium relevance — key rules only
|
|
32
|
+
├── Level 2 (Available): listed for reference only
|
|
33
|
+
└── Large models: Essential + Supplementary | Small models: Essential only
|
|
34
|
+
│
|
|
35
|
+
▼
|
|
36
|
+
Context scan (MANDATORY before first line of code)
|
|
37
|
+
├── Read package.json → verify all imports exist
|
|
38
|
+
├── Read tsconfig.json → understand strictness, paths aliases
|
|
39
|
+
├── Read referenced files → understand actual data shapes
|
|
40
|
+
└── Read .env.example → know available environment variables
|
|
41
|
+
│
|
|
42
|
+
▼
|
|
43
|
+
Maker generates at temperature 0.1
|
|
44
|
+
├── Only methods verified in official docs
|
|
45
|
+
├── Only packages in package.json
|
|
46
|
+
├── // VERIFY: [reason] on any uncertain call
|
|
47
|
+
└── No full application generation — modules only
|
|
48
|
+
│
|
|
49
|
+
▼
|
|
50
|
+
[Phase 6] Inner-Loop Validator (AUTO — runs before you see the code)
|
|
51
|
+
├── Scans generated snippet for OWASP patterns (critical/high/medium/low)
|
|
52
|
+
├── Runs structural heuristics (empty catch, throw strings, env without fallback)
|
|
53
|
+
├── Verdict: APPROVED / WARNING / REJECTED
|
|
54
|
+
│ ├── APPROVED → continues to Tribunal Review
|
|
55
|
+
│ ├── WARNING → noted, continues with flag
|
|
56
|
+
│ └── REJECTED → Maker auto-corrects (up to 2 inner-loop attempts)
|
|
57
|
+
└── Only clean code reaches the Tribunal reviewers
|
|
58
|
+
│
|
|
59
|
+
▼
|
|
60
|
+
Tribunal Reviewers run in parallel (auto-selected by keyword)
|
|
61
|
+
│
|
|
62
|
+
▼
|
|
63
|
+
Human Gate — verdicts shown + unified diff
|
|
64
|
+
Y = write to disk | N = discard | R = revise with feedback
|
|
65
|
+
```
|
|
66
|
+
|
|
67
|
+
---
|
|
68
|
+
|
|
69
|
+
## What the Maker Is Not Allowed to Do
|
|
70
|
+
|
|
71
|
+
```
|
|
72
|
+
❌ Import a package not in package.json
|
|
73
|
+
❌ Call a method not verified in official documentation
|
|
74
|
+
❌ Use TypeScript 'any' without an explanation comment
|
|
75
|
+
❌ Generate an entire application in one shot
|
|
76
|
+
❌ Guess at database column or table names (read schema first)
|
|
77
|
+
❌ Fabricate API response shapes (read existing types first)
|
|
78
|
+
❌ Assume environment variables exist (read .env.example first)
|
|
79
|
+
❌ Use Next.js 14 patterns in a Next.js 15 project (check version!)
|
|
80
|
+
❌ Use React 18 hooks in a React 19 project (useFormState → useActionState)
|
|
81
|
+
❌ Use framer-motion v6 API in a v12 project (exitBeforeEnter → mode="wait")
|
|
82
|
+
❌ Use raw useEffect for GSAP — always useGSAP from @gsap/react
|
|
83
|
+
❌ Hallucinate LLM model names — verify against provider's current model list
|
|
84
|
+
```
|
|
85
|
+
|
|
86
|
+
When unsure: write `// VERIFY: [specific reason]` instead of hallucinating.
|
|
87
|
+
|
|
88
|
+
---
|
|
89
|
+
|
|
90
|
+
## Reviewer Auto-Selection
|
|
91
|
+
|
|
92
|
+
**Always active:**
|
|
93
|
+
```
|
|
94
|
+
precedence-reviewer→ Enforces repository Case Law and past rejections (Runs First)
|
|
95
|
+
logic-reviewer → Hallucinated methods, undefined refs, impossible logic
|
|
96
|
+
security-auditor → OWASP vulnerabilities, hardcoded secrets, injection
|
|
97
|
+
```
|
|
98
|
+
|
|
99
|
+
**Auto-activated by keywords:**
|
|
100
|
+
|
|
101
|
+
| Keyword in request | Additional Reviewers |
|
|
102
|
+
|:---|:---|
|
|
103
|
+
| `api`, `route`, `endpoint`, `handler`, `server action` | `dependency-reviewer` + `type-safety-reviewer` |
|
|
104
|
+
| `sql`, `query`, `database`, `prisma`, `drizzle`, `orm` | `sql-reviewer` |
|
|
105
|
+
| `component`, `hook`, `react`, `vue`, `jsx`, `tsx` | `frontend-reviewer` + `type-safety-reviewer` + `ui-ux-auditor` |
|
|
106
|
+
| `ui`, `design`, `landing`, `page`, `layout`, `style`, `css` | `ui-ux-auditor` + `accessibility-reviewer` |
|
|
107
|
+
| `animation`, `gsap`, `framer`, `motion`, `scroll` | `frontend-reviewer` + `performance-reviewer` + `ui-ux-auditor` |
|
|
108
|
+
| `test`, `spec`, `vitest`, `jest`, `playwright` | `test-coverage-reviewer` |
|
|
109
|
+
| `slow`, `optimize`, `cache`, `performance`, `bundle` | `performance-reviewer` |
|
|
110
|
+
| `mobile`, `react native`, `expo` | `mobile-reviewer` |
|
|
111
|
+
| `llm`, `openai`, `anthropic`, `gemini`, `embedding`, `ai` | `ai-code-reviewer` |
|
|
112
|
+
| `aria`, `wcag`, `a11y`, `accessibility` | `accessibility-reviewer` + `ui-ux-auditor` |
|
|
113
|
+
| `import`, `package`, `npm`, `require` | `dependency-reviewer` |
|
|
114
|
+
|
|
115
|
+
> For maximum safety on critical code: use `/tribunal-full` for all 11 reviewers simultaneously.
|
|
116
|
+
|
|
117
|
+
---
|
|
118
|
+
|
|
119
|
+
## Reviewer Verdicts
|
|
120
|
+
|
|
121
|
+
| Verdict | Meaning | What Happens |
|
|
122
|
+
|:---|:---|:---|
|
|
123
|
+
| `✅ APPROVED` | No issues found | Proceeds to Human Gate |
|
|
124
|
+
| `⚠️ WARNING` | Non-blocking issue | Human Gate shown with warning highlighted |
|
|
125
|
+
| `❌ REJECTED` | Blocking issue | Maker revises before Human Gate |
|
|
126
|
+
|
|
127
|
+
**Retry limit:** Maker is revised up to 3 times per REJECTED verdict. After 3 failures, the session halts and reports to the user with full failure history. No silent failures.
|
|
128
|
+
|
|
129
|
+
---
|
|
130
|
+
|
|
131
|
+
## Output Format
|
|
132
|
+
|
|
133
|
+
```
|
|
134
|
+
━━━ Tribunal: [Domain] ━━━━━━━━━━━━━━━━━━━━━━
|
|
135
|
+
|
|
136
|
+
Active reviewers: logic · security · [others]
|
|
137
|
+
|
|
138
|
+
[Generated code with // VERIFY: tags where uncertain]
|
|
139
|
+
|
|
140
|
+
━━━ Verdicts ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
|
|
141
|
+
|
|
142
|
+
logic-reviewer: ✅ APPROVED
|
|
143
|
+
security-auditor: ✅ APPROVED
|
|
144
|
+
dependency-reviewer: ⚠️ WARNING — lodash not in package.json
|
|
145
|
+
|
|
146
|
+
━━━ Warnings ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
|
|
147
|
+
|
|
148
|
+
dependency-reviewer:
|
|
149
|
+
⚠️ Medium — Line 3: 'lodash' imported but not in package.json
|
|
150
|
+
Fix: npm install lodash OR use built-in Array methods
|
|
151
|
+
|
|
152
|
+
━━━ Human Gate ━━━━━━━━━━━━━━━━━━━━━━━━━━━━
|
|
153
|
+
Write to disk? Y = approve | N = discard | R = revise with feedback
|
|
154
|
+
```
|
|
155
|
+
|
|
156
|
+
---
|
|
157
|
+
|
|
158
|
+
| After /generate shows... | Go to |
|
|
159
|
+
|:---|:---|
|
|
160
|
+
| Multiple files need changing | `/enhance` for impact-zone analysis |
|
|
161
|
+
| Security-critical code was generated | `/tribunal-full` for maximum coverage |
|
|
162
|
+
| DB queries were generated | `/tribunal-database` |
|
|
163
|
+
| New API routes were generated | `/tribunal-backend` |
|
|
164
|
+
| Animation/motion code generated | `/tribunal-frontend` |
|
|
165
|
+
| Tests need to be written next | `/test` |
|
|
166
|
+
| Something was rejected 3 times | Escalate to human with failure report |
|
|
167
|
+
|
|
168
|
+
---
|
|
169
|
+
|
|
170
|
+
## Usage Examples
|
|
171
|
+
|
|
172
|
+
```
|
|
173
|
+
/generate a JWT middleware for Express with HS256 algorithm enforcement
|
|
174
|
+
/generate a Prisma query for users with their published posts in last 30 days
|
|
175
|
+
/generate a debounced search hook in React 19 using useDeferredValue
|
|
176
|
+
/generate a parameterized SQL query for paginated order history
|
|
177
|
+
/generate a Zod schema for email + password + role login input
|
|
178
|
+
/generate a Server Action for creating a product with image upload
|
|
179
|
+
/generate a rate-limited fetch wrapper using @upstash/ratelimit
|
|
180
|
+
/generate a Framer Motion page transition with shared element (layoutId)
|
|
181
|
+
/generate a GSAP ScrollTrigger timeline with useGSAP React hook
|
|
182
|
+
/generate an OpenAI structured output call with Zod schema validation
|
|
183
|
+
```
|