jorgex-stack 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.
Files changed (124) hide show
  1. package/LICENSE +21 -0
  2. package/PRD.md +297 -0
  3. package/README.md +58 -0
  4. package/dist/cli.js +2894 -0
  5. package/package.json +51 -0
  6. package/stack/agents/README.md +33 -0
  7. package/stack/agents/backend-analyst.md +61 -0
  8. package/stack/agents/code-reviewer.md +66 -0
  9. package/stack/agents/code-simplifier.md +81 -0
  10. package/stack/agents/comment-fixer.md +54 -0
  11. package/stack/agents/docs-maintainer.md +88 -0
  12. package/stack/agents/engram.md +124 -0
  13. package/stack/agents/frontend-analyst.md +51 -0
  14. package/stack/agents/implementer.md +55 -0
  15. package/stack/agents/orchestrator.md +192 -0
  16. package/stack/agents/security-auditor.md +66 -0
  17. package/stack/agents/silent-failure-hunter.md +167 -0
  18. package/stack/agents/test-analyzer.md +91 -0
  19. package/stack/agents/tester.md +71 -0
  20. package/stack/agents/translator.md +101 -0
  21. package/stack/agents/type-design-analyzer.md +128 -0
  22. package/stack/commands/xreview.md +80 -0
  23. package/stack/config/defaults.json +37 -0
  24. package/stack/hooks/hooks.json +18 -0
  25. package/stack/mcp/servers.json +19 -0
  26. package/stack/plugins/opencode/engram.ts +378 -0
  27. package/stack/plugins/opencode/hooks.ts +766 -0
  28. package/stack/plugins/opencode/package.json +10 -0
  29. package/stack/plugins/opencode/worktree.ts +405 -0
  30. package/stack/scripts/post-pr-review.cjs +156 -0
  31. package/stack/skills/agent-browser/SKILL.md +55 -0
  32. package/stack/skills/agent-delegation/SKILL.md +58 -0
  33. package/stack/skills/deploy-to-vercel/SKILL.md +296 -0
  34. package/stack/skills/deploy-to-vercel/resources/deploy-codex.sh +301 -0
  35. package/stack/skills/deploy-to-vercel/resources/deploy.sh +301 -0
  36. package/stack/skills/diagnose/SKILL.md +117 -0
  37. package/stack/skills/diagnose/scripts/hitl-loop.template.sh +41 -0
  38. package/stack/skills/find-skills/SKILL.md +133 -0
  39. package/stack/skills/graphify/.graphify_version +1 -0
  40. package/stack/skills/graphify/SKILL.md +1319 -0
  41. package/stack/skills/mcp-builder/LICENSE.txt +202 -0
  42. package/stack/skills/mcp-builder/SKILL.md +236 -0
  43. package/stack/skills/mcp-builder/reference/evaluation.md +602 -0
  44. package/stack/skills/mcp-builder/reference/mcp_best_practices.md +249 -0
  45. package/stack/skills/mcp-builder/reference/node_mcp_server.md +970 -0
  46. package/stack/skills/mcp-builder/reference/python_mcp_server.md +719 -0
  47. package/stack/skills/mcp-builder/scripts/connections.py +151 -0
  48. package/stack/skills/mcp-builder/scripts/evaluation.py +373 -0
  49. package/stack/skills/mcp-builder/scripts/example_evaluation.xml +22 -0
  50. package/stack/skills/mcp-builder/scripts/requirements.txt +2 -0
  51. package/stack/skills/obsidian-cli/SKILL.md +106 -0
  52. package/stack/skills/obsidian-markdown/SKILL.md +196 -0
  53. package/stack/skills/obsidian-markdown/references/CALLOUTS.md +58 -0
  54. package/stack/skills/obsidian-markdown/references/EMBEDS.md +63 -0
  55. package/stack/skills/obsidian-markdown/references/PROPERTIES.md +61 -0
  56. package/stack/skills/react-doctor/SKILL.md +19 -0
  57. package/stack/skills/skill-creator/LICENSE.txt +202 -0
  58. package/stack/skills/skill-creator/SKILL.md +485 -0
  59. package/stack/skills/skill-creator/agents/analyzer.md +274 -0
  60. package/stack/skills/skill-creator/agents/comparator.md +202 -0
  61. package/stack/skills/skill-creator/agents/grader.md +223 -0
  62. package/stack/skills/skill-creator/assets/eval_review.html +146 -0
  63. package/stack/skills/skill-creator/eval-viewer/generate_review.py +471 -0
  64. package/stack/skills/skill-creator/eval-viewer/viewer.html +1325 -0
  65. package/stack/skills/skill-creator/references/schemas.md +430 -0
  66. package/stack/skills/skill-creator/scripts/aggregate_benchmark.py +401 -0
  67. package/stack/skills/skill-creator/scripts/generate_report.py +326 -0
  68. package/stack/skills/skill-creator/scripts/improve_description.py +248 -0
  69. package/stack/skills/skill-creator/scripts/package_skill.py +136 -0
  70. package/stack/skills/skill-creator/scripts/quick_validate.py +103 -0
  71. package/stack/skills/skill-creator/scripts/run_eval.py +310 -0
  72. package/stack/skills/skill-creator/scripts/run_loop.py +332 -0
  73. package/stack/skills/skill-creator/scripts/utils.py +47 -0
  74. package/stack/skills/supabase/SKILL.md +135 -0
  75. package/stack/skills/supabase/assets/feedback-issue-template.md +17 -0
  76. package/stack/skills/supabase/references/skill-feedback.md +17 -0
  77. package/stack/skills/supabase-postgres-best-practices/SKILL.md +64 -0
  78. package/stack/skills/supabase-postgres-best-practices/references/_contributing.md +170 -0
  79. package/stack/skills/supabase-postgres-best-practices/references/_sections.md +39 -0
  80. package/stack/skills/supabase-postgres-best-practices/references/_template.md +34 -0
  81. package/stack/skills/supabase-postgres-best-practices/references/advanced-full-text-search.md +55 -0
  82. package/stack/skills/supabase-postgres-best-practices/references/advanced-jsonb-indexing.md +49 -0
  83. package/stack/skills/supabase-postgres-best-practices/references/conn-idle-timeout.md +46 -0
  84. package/stack/skills/supabase-postgres-best-practices/references/conn-limits.md +44 -0
  85. package/stack/skills/supabase-postgres-best-practices/references/conn-pooling.md +41 -0
  86. package/stack/skills/supabase-postgres-best-practices/references/conn-prepared-statements.md +46 -0
  87. package/stack/skills/supabase-postgres-best-practices/references/data-batch-inserts.md +54 -0
  88. package/stack/skills/supabase-postgres-best-practices/references/data-n-plus-one.md +53 -0
  89. package/stack/skills/supabase-postgres-best-practices/references/data-pagination.md +50 -0
  90. package/stack/skills/supabase-postgres-best-practices/references/data-upsert.md +50 -0
  91. package/stack/skills/supabase-postgres-best-practices/references/lock-advisory.md +56 -0
  92. package/stack/skills/supabase-postgres-best-practices/references/lock-deadlock-prevention.md +68 -0
  93. package/stack/skills/supabase-postgres-best-practices/references/lock-short-transactions.md +50 -0
  94. package/stack/skills/supabase-postgres-best-practices/references/lock-skip-locked.md +54 -0
  95. package/stack/skills/supabase-postgres-best-practices/references/monitor-explain-analyze.md +45 -0
  96. package/stack/skills/supabase-postgres-best-practices/references/monitor-pg-stat-statements.md +55 -0
  97. package/stack/skills/supabase-postgres-best-practices/references/monitor-vacuum-analyze.md +55 -0
  98. package/stack/skills/supabase-postgres-best-practices/references/query-composite-indexes.md +44 -0
  99. package/stack/skills/supabase-postgres-best-practices/references/query-covering-indexes.md +40 -0
  100. package/stack/skills/supabase-postgres-best-practices/references/query-index-types.md +48 -0
  101. package/stack/skills/supabase-postgres-best-practices/references/query-missing-indexes.md +43 -0
  102. package/stack/skills/supabase-postgres-best-practices/references/query-partial-indexes.md +45 -0
  103. package/stack/skills/supabase-postgres-best-practices/references/schema-constraints.md +80 -0
  104. package/stack/skills/supabase-postgres-best-practices/references/schema-data-types.md +46 -0
  105. package/stack/skills/supabase-postgres-best-practices/references/schema-foreign-key-indexes.md +59 -0
  106. package/stack/skills/supabase-postgres-best-practices/references/schema-lowercase-identifiers.md +55 -0
  107. package/stack/skills/supabase-postgres-best-practices/references/schema-partitioning.md +55 -0
  108. package/stack/skills/supabase-postgres-best-practices/references/schema-primary-keys.md +61 -0
  109. package/stack/skills/supabase-postgres-best-practices/references/security-privileges.md +54 -0
  110. package/stack/skills/supabase-postgres-best-practices/references/security-rls-basics.md +50 -0
  111. package/stack/skills/supabase-postgres-best-practices/references/security-rls-performance.md +63 -0
  112. package/stack/skills/tdd/SKILL.md +109 -0
  113. package/stack/skills/tdd/deep-modules.md +33 -0
  114. package/stack/skills/tdd/interface-design.md +31 -0
  115. package/stack/skills/tdd/mocking.md +59 -0
  116. package/stack/skills/tdd/refactoring.md +10 -0
  117. package/stack/skills/tdd/tests.md +61 -0
  118. package/stack/skills/to-issues/SKILL.md +83 -0
  119. package/stack/skills/to-prd/SKILL.md +72 -0
  120. package/stack/skills/work-lifecycle/SKILL.md +79 -0
  121. package/stack/skills/work-lifecycle/references/plan-template.md +185 -0
  122. package/stack/system-prompt/AGENTS.md +171 -0
  123. package/stack/system-prompt/engram-protocol.md +53 -0
  124. package/upstreams.json +96 -0
@@ -0,0 +1,301 @@
1
+ #!/bin/bash
2
+
3
+ # Vercel Deployment Script (via claimable deploy endpoint)
4
+ # Usage: ./deploy.sh [project-path]
5
+ # Returns: JSON with previewUrl, claimUrl, deploymentId, projectId
6
+
7
+ set -euo pipefail
8
+
9
+ DEPLOY_ENDPOINT="https://claude-skills-deploy.vercel.com/api/deploy"
10
+
11
+ # Detect framework from package.json
12
+ detect_framework() {
13
+ local pkg_json="$1"
14
+
15
+ if [ ! -f "$pkg_json" ]; then
16
+ echo "null"
17
+ return
18
+ fi
19
+
20
+ local content=$(cat "$pkg_json")
21
+
22
+ # Helper to check if a package exists in dependencies or devDependencies.
23
+ # Use exact matching by default, with a separate prefix matcher for scoped
24
+ # package families like "@remix-run/".
25
+ has_dep_exact() {
26
+ echo "$content" | grep -q "\"$1\""
27
+ }
28
+
29
+ has_dep_prefix() {
30
+ echo "$content" | grep -q "\"$1"
31
+ }
32
+
33
+ # Order matters - check more specific frameworks first
34
+
35
+ # Blitz
36
+ if has_dep_exact "blitz"; then echo "blitzjs"; return; fi
37
+
38
+ # Next.js
39
+ if has_dep_exact "next"; then echo "nextjs"; return; fi
40
+
41
+ # Gatsby
42
+ if has_dep_exact "gatsby"; then echo "gatsby"; return; fi
43
+
44
+ # Remix
45
+ if has_dep_prefix "@remix-run/"; then echo "remix"; return; fi
46
+
47
+ # React Router (v7 framework mode)
48
+ if has_dep_prefix "@react-router/"; then echo "react-router"; return; fi
49
+
50
+ # TanStack Start
51
+ if has_dep_exact "@tanstack/start"; then echo "tanstack-start"; return; fi
52
+
53
+ # Astro
54
+ if has_dep_exact "astro"; then echo "astro"; return; fi
55
+
56
+ # Hydrogen (Shopify)
57
+ if has_dep_exact "@shopify/hydrogen"; then echo "hydrogen"; return; fi
58
+
59
+ # SvelteKit
60
+ if has_dep_exact "@sveltejs/kit"; then echo "sveltekit-1"; return; fi
61
+
62
+ # Svelte (standalone)
63
+ if has_dep_exact "svelte"; then echo "svelte"; return; fi
64
+
65
+ # Nuxt
66
+ if has_dep_exact "nuxt"; then echo "nuxtjs"; return; fi
67
+
68
+ # Vue with Vitepress
69
+ if has_dep_exact "vitepress"; then echo "vitepress"; return; fi
70
+
71
+ # Vue with Vuepress
72
+ if has_dep_exact "vuepress"; then echo "vuepress"; return; fi
73
+
74
+ # Gridsome
75
+ if has_dep_exact "gridsome"; then echo "gridsome"; return; fi
76
+
77
+ # SolidStart
78
+ if has_dep_exact "@solidjs/start"; then echo "solidstart-1"; return; fi
79
+
80
+ # Docusaurus
81
+ if has_dep_exact "@docusaurus/core"; then echo "docusaurus-2"; return; fi
82
+
83
+ # RedwoodJS
84
+ if has_dep_prefix "@redwoodjs/"; then echo "redwoodjs"; return; fi
85
+
86
+ # Hexo
87
+ if has_dep_exact "hexo"; then echo "hexo"; return; fi
88
+
89
+ # Eleventy
90
+ if has_dep_exact "@11ty/eleventy"; then echo "eleventy"; return; fi
91
+
92
+ # Angular / Ionic Angular
93
+ if has_dep_exact "@ionic/angular"; then echo "ionic-angular"; return; fi
94
+ if has_dep_exact "@angular/core"; then echo "angular"; return; fi
95
+
96
+ # Ionic React
97
+ if has_dep_exact "@ionic/react"; then echo "ionic-react"; return; fi
98
+
99
+ # Create React App
100
+ if has_dep_exact "react-scripts"; then echo "create-react-app"; return; fi
101
+
102
+ # Ember
103
+ if has_dep_exact "ember-cli" || has_dep_exact "ember-source"; then echo "ember"; return; fi
104
+
105
+ # Dojo
106
+ if has_dep_exact "@dojo/framework"; then echo "dojo"; return; fi
107
+
108
+ # Polymer
109
+ if has_dep_prefix "@polymer/"; then echo "polymer"; return; fi
110
+
111
+ # Preact
112
+ if has_dep_exact "preact"; then echo "preact"; return; fi
113
+
114
+ # Stencil
115
+ if has_dep_exact "@stencil/core"; then echo "stencil"; return; fi
116
+
117
+ # UmiJS
118
+ if has_dep_exact "umi"; then echo "umijs"; return; fi
119
+
120
+ # Sapper (legacy Svelte)
121
+ if has_dep_exact "sapper"; then echo "sapper"; return; fi
122
+
123
+ # Saber
124
+ if has_dep_exact "saber"; then echo "saber"; return; fi
125
+
126
+ # Sanity
127
+ if has_dep_exact "sanity"; then echo "sanity-v3"; return; fi
128
+ if has_dep_prefix "@sanity/"; then echo "sanity"; return; fi
129
+
130
+ # Storybook
131
+ if has_dep_prefix "@storybook/"; then echo "storybook"; return; fi
132
+
133
+ # NestJS
134
+ if has_dep_exact "@nestjs/core"; then echo "nestjs"; return; fi
135
+
136
+ # Elysia
137
+ if has_dep_exact "elysia"; then echo "elysia"; return; fi
138
+
139
+ # Hono
140
+ if has_dep_exact "hono"; then echo "hono"; return; fi
141
+
142
+ # Fastify
143
+ if has_dep_exact "fastify"; then echo "fastify"; return; fi
144
+
145
+ # h3
146
+ if has_dep_exact "h3"; then echo "h3"; return; fi
147
+
148
+ # Nitro
149
+ if has_dep_exact "nitropack"; then echo "nitro"; return; fi
150
+
151
+ # Express
152
+ if has_dep_exact "express"; then echo "express"; return; fi
153
+
154
+ # Vite (generic - check last among JS frameworks)
155
+ if has_dep_exact "vite"; then echo "vite"; return; fi
156
+
157
+ # Parcel
158
+ if has_dep_exact "parcel"; then echo "parcel"; return; fi
159
+
160
+ # No framework detected
161
+ echo "null"
162
+ }
163
+
164
+ # Parse arguments
165
+ INPUT_PATH="${1:-.}"
166
+
167
+ # Create temp directory for packaging
168
+ TEMP_DIR=$(mktemp -d)
169
+ TARBALL="$TEMP_DIR/project.tgz"
170
+ STAGING_DIR="$TEMP_DIR/staging"
171
+ CLEANUP_TEMP=true
172
+
173
+ cleanup() {
174
+ if [ "$CLEANUP_TEMP" = true ]; then
175
+ rm -rf "$TEMP_DIR"
176
+ fi
177
+ }
178
+ trap cleanup EXIT
179
+
180
+ echo "Preparing deployment..." >&2
181
+
182
+ # Check if input is a .tgz file or a directory
183
+ FRAMEWORK="null"
184
+
185
+ if [ -f "$INPUT_PATH" ] && [[ "$INPUT_PATH" == *.tgz ]]; then
186
+ # Input is already a tarball, use it directly
187
+ echo "Using provided tarball..." >&2
188
+ TARBALL="$INPUT_PATH"
189
+ CLEANUP_TEMP=false
190
+ # Can't detect framework from tarball, leave as null
191
+ elif [ -d "$INPUT_PATH" ]; then
192
+ # Input is a directory, need to tar it
193
+ PROJECT_PATH=$(cd "$INPUT_PATH" && pwd)
194
+
195
+ # Detect framework from package.json
196
+ FRAMEWORK=$(detect_framework "$PROJECT_PATH/package.json")
197
+
198
+ # Stage files into a temporary directory to avoid mutating the source tree.
199
+ mkdir -p "$STAGING_DIR"
200
+ echo "Staging project files..." >&2
201
+ tar -C "$PROJECT_PATH" \
202
+ --exclude='node_modules' \
203
+ --exclude='.git' \
204
+ --exclude='.env' \
205
+ --exclude='.env.*' \
206
+ -cf - . | tar -C "$STAGING_DIR" -xf -
207
+
208
+ # Check if this is a static HTML project (no package.json)
209
+ if [ ! -f "$PROJECT_PATH/package.json" ]; then
210
+ # Find HTML files in root
211
+ HTML_FILES=$(find "$STAGING_DIR" -maxdepth 1 -name "*.html" -type f)
212
+ HTML_COUNT=$(printf '%s\n' "$HTML_FILES" | sed '/^$/d' | wc -l | tr -d '[:space:]')
213
+
214
+ # If there's exactly one HTML file and it's not index.html, rename it
215
+ if [ "$HTML_COUNT" -eq 1 ]; then
216
+ HTML_FILE=$(echo "$HTML_FILES" | head -1)
217
+ BASENAME=$(basename "$HTML_FILE")
218
+ if [ "$BASENAME" != "index.html" ]; then
219
+ echo "Renaming $BASENAME to index.html..." >&2
220
+ mv "$HTML_FILE" "$STAGING_DIR/index.html"
221
+ fi
222
+ fi
223
+ fi
224
+
225
+ # Create tarball from the staging directory
226
+ echo "Creating deployment package..." >&2
227
+ tar -czf "$TARBALL" -C "$STAGING_DIR" .
228
+ else
229
+ echo "Error: Input must be a directory or a .tgz file" >&2
230
+ exit 1
231
+ fi
232
+
233
+ if [ "$FRAMEWORK" != "null" ]; then
234
+ echo "Detected framework: $FRAMEWORK" >&2
235
+ fi
236
+
237
+ # Deploy
238
+ echo "Deploying..." >&2
239
+ RESPONSE=$(curl -s -X POST "$DEPLOY_ENDPOINT" -F "file=@$TARBALL" -F "framework=$FRAMEWORK")
240
+
241
+ # Check for error in response
242
+ if echo "$RESPONSE" | grep -q '"error"'; then
243
+ ERROR_MSG=$(echo "$RESPONSE" | grep -o '"error":"[^"]*"' | cut -d'"' -f4)
244
+ echo "Error: $ERROR_MSG" >&2
245
+ exit 1
246
+ fi
247
+
248
+ # Extract URLs from response
249
+ PREVIEW_URL=$(echo "$RESPONSE" | grep -o '"previewUrl":"[^"]*"' | cut -d'"' -f4)
250
+ CLAIM_URL=$(echo "$RESPONSE" | grep -o '"claimUrl":"[^"]*"' | cut -d'"' -f4)
251
+
252
+ if [ -z "$PREVIEW_URL" ]; then
253
+ echo "Error: Could not extract preview URL from response" >&2
254
+ echo "$RESPONSE" >&2
255
+ exit 1
256
+ fi
257
+
258
+ echo "Deployment started. Waiting for build to complete..." >&2
259
+ echo "Preview URL: $PREVIEW_URL" >&2
260
+
261
+ # Poll the preview URL until it returns a non-5xx response (5xx = still building)
262
+ MAX_ATTEMPTS=60 # 5 minutes max (60 * 5 seconds)
263
+ ATTEMPT=0
264
+
265
+ while [ $ATTEMPT -lt $MAX_ATTEMPTS ]; do
266
+ HTTP_STATUS=$(curl -s -o /dev/null -w "%{http_code}" "$PREVIEW_URL")
267
+
268
+ if [ "$HTTP_STATUS" -eq 200 ]; then
269
+ echo "" >&2
270
+ echo "Deployment ready!" >&2
271
+ break
272
+ elif [ "$HTTP_STATUS" -ge 500 ]; then
273
+ # 5xx means still building/deploying
274
+ echo "Building... (attempt $((ATTEMPT + 1))/$MAX_ATTEMPTS)" >&2
275
+ sleep 5
276
+ ATTEMPT=$((ATTEMPT + 1))
277
+ elif [ "$HTTP_STATUS" -ge 400 ] && [ "$HTTP_STATUS" -lt 500 ]; then
278
+ # 4xx might be an error or the app itself returns 4xx - it's responding
279
+ echo "" >&2
280
+ echo "Deployment ready (returned $HTTP_STATUS)!" >&2
281
+ break
282
+ else
283
+ # Any other status, assume it's ready
284
+ echo "" >&2
285
+ echo "Deployment ready!" >&2
286
+ break
287
+ fi
288
+ done
289
+
290
+ if [ $ATTEMPT -eq $MAX_ATTEMPTS ]; then
291
+ echo "" >&2
292
+ echo "Warning: Timed out waiting for deployment, but it may still be building." >&2
293
+ fi
294
+
295
+ echo "" >&2
296
+ echo "Preview URL: $PREVIEW_URL" >&2
297
+ echo "Claim URL: $CLAIM_URL" >&2
298
+ echo "" >&2
299
+
300
+ # Output JSON for programmatic use
301
+ echo "$RESPONSE"
@@ -0,0 +1,117 @@
1
+ ---
2
+ name: diagnose
3
+ description: Disciplined diagnosis loop for hard bugs and performance regressions. Reproduce → minimise → hypothesise → instrument → fix → regression-test. Use when user says "diagnose this" / "debug this", reports a bug, says something is broken/throwing/failing, or describes a performance regression.
4
+ ---
5
+
6
+ # Diagnose
7
+
8
+ A discipline for hard bugs. Skip phases only when explicitly justified.
9
+
10
+ When exploring the codebase, use the project's domain glossary to get a clear mental model of the relevant modules, and check ADRs in the area you're touching.
11
+
12
+ ## Phase 1 — Build a feedback loop
13
+
14
+ **This is the skill.** Everything else is mechanical. If you have a fast, deterministic, agent-runnable pass/fail signal for the bug, you will find the cause — bisection, hypothesis-testing, and instrumentation all just consume that signal. If you don't have one, no amount of staring at code will save you.
15
+
16
+ Spend disproportionate effort here. **Be aggressive. Be creative. Refuse to give up.**
17
+
18
+ ### Ways to construct one — try them in roughly this order
19
+
20
+ 1. **Failing test** at whatever seam reaches the bug — unit, integration, e2e.
21
+ 2. **Curl / HTTP script** against a running dev server.
22
+ 3. **CLI invocation** with a fixture input, diffing stdout against a known-good snapshot.
23
+ 4. **Headless browser script** (Playwright / Puppeteer) — drives the UI, asserts on DOM/console/network.
24
+ 5. **Replay a captured trace.** Save a real network request / payload / event log to disk; replay it through the code path in isolation.
25
+ 6. **Throwaway harness.** Spin up a minimal subset of the system (one service, mocked deps) that exercises the bug code path with a single function call.
26
+ 7. **Property / fuzz loop.** If the bug is "sometimes wrong output", run 1000 random inputs and look for the failure mode.
27
+ 8. **Bisection harness.** If the bug appeared between two known states (commit, dataset, version), automate "boot at state X, check, repeat" so you can `git bisect run` it.
28
+ 9. **Differential loop.** Run the same input through old-version vs new-version (or two configs) and diff outputs.
29
+ 10. **HITL bash script.** Last resort. If a human must click, drive _them_ with `scripts/hitl-loop.template.sh` so the loop is still structured. Captured output feeds back to you.
30
+
31
+ Build the right feedback loop, and the bug is 90% fixed.
32
+
33
+ ### Iterate on the loop itself
34
+
35
+ Treat the loop as a product. Once you have _a_ loop, ask:
36
+
37
+ - Can I make it faster? (Cache setup, skip unrelated init, narrow the test scope.)
38
+ - Can I make the signal sharper? (Assert on the specific symptom, not "didn't crash".)
39
+ - Can I make it more deterministic? (Pin time, seed RNG, isolate filesystem, freeze network.)
40
+
41
+ A 30-second flaky loop is barely better than no loop. A 2-second deterministic loop is a debugging superpower.
42
+
43
+ ### Non-deterministic bugs
44
+
45
+ The goal is not a clean repro but a **higher reproduction rate**. Loop the trigger 100×, parallelise, add stress, narrow timing windows, inject sleeps. A 50%-flake bug is debuggable; 1% is not — keep raising the rate until it's debuggable.
46
+
47
+ ### When you genuinely cannot build a loop
48
+
49
+ Stop and say so explicitly. List what you tried. Ask the user for: (a) access to whatever environment reproduces it, (b) a captured artifact (HAR file, log dump, core dump, screen recording with timestamps), or (c) permission to add temporary production instrumentation. Do **not** proceed to hypothesise without a loop.
50
+
51
+ Do not proceed to Phase 2 until you have a loop you believe in.
52
+
53
+ ## Phase 2 — Reproduce
54
+
55
+ Run the loop. Watch the bug appear.
56
+
57
+ Confirm:
58
+
59
+ - [ ] The loop produces the failure mode the **user** described — not a different failure that happens to be nearby. Wrong bug = wrong fix.
60
+ - [ ] The failure is reproducible across multiple runs (or, for non-deterministic bugs, reproducible at a high enough rate to debug against).
61
+ - [ ] You have captured the exact symptom (error message, wrong output, slow timing) so later phases can verify the fix actually addresses it.
62
+
63
+ Do not proceed until you reproduce the bug.
64
+
65
+ ## Phase 3 — Hypothesise
66
+
67
+ Generate **3–5 ranked hypotheses** before testing any of them. Single-hypothesis generation anchors on the first plausible idea.
68
+
69
+ Each hypothesis must be **falsifiable**: state the prediction it makes.
70
+
71
+ > Format: "If <X> is the cause, then <changing Y> will make the bug disappear / <changing Z> will make it worse."
72
+
73
+ If you cannot state the prediction, the hypothesis is a vibe — discard or sharpen it.
74
+
75
+ **Show the ranked list to the user before testing.** They often have domain knowledge that re-ranks instantly ("we just deployed a change to #3"), or know hypotheses they've already ruled out. Cheap checkpoint, big time saver. Don't block on it — proceed with your ranking if the user is AFK.
76
+
77
+ ## Phase 4 — Instrument
78
+
79
+ Each probe must map to a specific prediction from Phase 3. **Change one variable at a time.**
80
+
81
+ Tool preference:
82
+
83
+ 1. **Debugger / REPL inspection** if the env supports it. One breakpoint beats ten logs.
84
+ 2. **Targeted logs** at the boundaries that distinguish hypotheses.
85
+ 3. Never "log everything and grep".
86
+
87
+ **Tag every debug log** with a unique prefix, e.g. `[DEBUG-a4f2]`. Cleanup at the end becomes a single grep. Untagged logs survive; tagged logs die.
88
+
89
+ **Perf branch.** For performance regressions, logs are usually wrong. Instead: establish a baseline measurement (timing harness, `performance.now()`, profiler, query plan), then bisect. Measure first, fix second.
90
+
91
+ ## Phase 5 — Fix + regression test
92
+
93
+ Write the regression test **before the fix** — but only if there is a **correct seam** for it.
94
+
95
+ A correct seam is one where the test exercises the **real bug pattern** as it occurs at the call site. If the only available seam is too shallow (single-caller test when the bug needs multiple callers, unit test that can't replicate the chain that triggered the bug), a regression test there gives false confidence.
96
+
97
+ **If no correct seam exists, that itself is the finding.** Note it. The codebase architecture is preventing the bug from being locked down. Flag this for the next phase.
98
+
99
+ If a correct seam exists:
100
+
101
+ 1. Turn the minimised repro into a failing test at that seam.
102
+ 2. Watch it fail.
103
+ 3. Apply the fix.
104
+ 4. Watch it pass.
105
+ 5. Re-run the Phase 1 feedback loop against the original (un-minimised) scenario.
106
+
107
+ ## Phase 6 — Cleanup + post-mortem
108
+
109
+ Required before declaring done:
110
+
111
+ - [ ] Original repro no longer reproduces (re-run the Phase 1 loop)
112
+ - [ ] Regression test passes (or absence of seam is documented)
113
+ - [ ] All `[DEBUG-...]` instrumentation removed (`grep` the prefix)
114
+ - [ ] Throwaway prototypes deleted (or moved to a clearly-marked debug location)
115
+ - [ ] The hypothesis that turned out correct is stated in the commit / PR message — so the next debugger learns
116
+
117
+ **Then ask: what would have prevented this bug?** If the answer involves architectural change (no good test seam, tangled callers, hidden coupling) hand off to the `/improve-codebase-architecture` skill with the specifics. Make the recommendation **after** the fix is in, not before — you have more information now than when you started.
@@ -0,0 +1,41 @@
1
+ #!/usr/bin/env bash
2
+ # Human-in-the-loop reproduction loop.
3
+ # Copy this file, edit the steps below, and run it.
4
+ # The agent runs the script; the user follows prompts in their terminal.
5
+ #
6
+ # Usage:
7
+ # bash hitl-loop.template.sh
8
+ #
9
+ # Two helpers:
10
+ # step "<instruction>" → show instruction, wait for Enter
11
+ # capture VAR "<question>" → show question, read response into VAR
12
+ #
13
+ # At the end, captured values are printed as KEY=VALUE for the agent to parse.
14
+
15
+ set -euo pipefail
16
+
17
+ step() {
18
+ printf '\n>>> %s\n' "$1"
19
+ read -r -p " [Enter when done] " _
20
+ }
21
+
22
+ capture() {
23
+ local var="$1" question="$2" answer
24
+ printf '\n>>> %s\n' "$question"
25
+ read -r -p " > " answer
26
+ printf -v "$var" '%s' "$answer"
27
+ }
28
+
29
+ # --- edit below ---------------------------------------------------------
30
+
31
+ step "Open the app at http://localhost:3000 and sign in."
32
+
33
+ capture ERRORED "Click the 'Export' button. Did it throw an error? (y/n)"
34
+
35
+ capture ERROR_MSG "Paste the error message (or 'none'):"
36
+
37
+ # --- edit above ---------------------------------------------------------
38
+
39
+ printf '\n--- Captured ---\n'
40
+ printf 'ERRORED=%s\n' "$ERRORED"
41
+ printf 'ERROR_MSG=%s\n' "$ERROR_MSG"
@@ -0,0 +1,133 @@
1
+ ---
2
+ name: find-skills
3
+ description: Helps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skill should be used when the user is looking for functionality that might exist as an installable skill.
4
+ ---
5
+
6
+ # Find Skills
7
+
8
+ This skill helps you discover and install skills from the open agent skills ecosystem.
9
+
10
+ ## When to Use This Skill
11
+
12
+ Use this skill when the user:
13
+
14
+ - Asks "how do I do X" where X might be a common task with an existing skill
15
+ - Says "find a skill for X" or "is there a skill for X"
16
+ - Asks "can you do X" where X is a specialized capability
17
+ - Expresses interest in extending agent capabilities
18
+ - Wants to search for tools, templates, or workflows
19
+ - Mentions they wish they had help with a specific domain (design, testing, deployment, etc.)
20
+
21
+ ## What is the Skills CLI?
22
+
23
+ The Skills CLI (`npx skills`) is the package manager for the open agent skills ecosystem. Skills are modular packages that extend agent capabilities with specialized knowledge, workflows, and tools.
24
+
25
+ **Key commands:**
26
+
27
+ - `npx skills find [query]` - Search for skills interactively or by keyword
28
+ - `npx skills add <package>` - Install a skill from GitHub or other sources
29
+ - `npx skills check` - Check for skill updates
30
+ - `npx skills update` - Update all installed skills
31
+
32
+ **Browse skills at:** https://skills.sh/
33
+
34
+ ## How to Help Users Find Skills
35
+
36
+ ### Step 1: Understand What They Need
37
+
38
+ When a user asks for help with something, identify:
39
+
40
+ 1. The domain (e.g., React, testing, design, deployment)
41
+ 2. The specific task (e.g., writing tests, creating animations, reviewing PRs)
42
+ 3. Whether this is a common enough task that a skill likely exists
43
+
44
+ ### Step 2: Search for Skills
45
+
46
+ Run the find command with a relevant query:
47
+
48
+ ```bash
49
+ npx skills find [query]
50
+ ```
51
+
52
+ For example:
53
+
54
+ - User asks "how do I make my React app faster?" → `npx skills find react performance`
55
+ - User asks "can you help me with PR reviews?" → `npx skills find pr review`
56
+ - User asks "I need to create a changelog" → `npx skills find changelog`
57
+
58
+ The command will return results like:
59
+
60
+ ```
61
+ Install with npx skills add <owner/repo@skill>
62
+
63
+ vercel-labs/agent-skills@vercel-react-best-practices
64
+ └ https://skills.sh/vercel-labs/agent-skills/vercel-react-best-practices
65
+ ```
66
+
67
+ ### Step 3: Present Options to the User
68
+
69
+ When you find relevant skills, present them to the user with:
70
+
71
+ 1. The skill name and what it does
72
+ 2. The install command they can run
73
+ 3. A link to learn more at skills.sh
74
+
75
+ Example response:
76
+
77
+ ```
78
+ I found a skill that might help! The "vercel-react-best-practices" skill provides
79
+ React and Next.js performance optimization guidelines from Vercel Engineering.
80
+
81
+ To install it:
82
+ npx skills add vercel-labs/agent-skills@vercel-react-best-practices
83
+
84
+ Learn more: https://skills.sh/vercel-labs/agent-skills/vercel-react-best-practices
85
+ ```
86
+
87
+ ### Step 4: Offer to Install
88
+
89
+ If the user wants to proceed, you can install the skill for them:
90
+
91
+ ```bash
92
+ npx skills add <owner/repo@skill> -g -y
93
+ ```
94
+
95
+ The `-g` flag installs globally (user-level) and `-y` skips confirmation prompts.
96
+
97
+ ## Common Skill Categories
98
+
99
+ When searching, consider these common categories:
100
+
101
+ | Category | Example Queries |
102
+ | --------------- | ---------------------------------------- |
103
+ | Web Development | react, nextjs, typescript, css, tailwind |
104
+ | Testing | testing, jest, playwright, e2e |
105
+ | DevOps | deploy, docker, kubernetes, ci-cd |
106
+ | Documentation | docs, readme, changelog, api-docs |
107
+ | Code Quality | review, lint, refactor, best-practices |
108
+ | Design | ui, ux, design-system, accessibility |
109
+ | Productivity | workflow, automation, git |
110
+
111
+ ## Tips for Effective Searches
112
+
113
+ 1. **Use specific keywords**: "react testing" is better than just "testing"
114
+ 2. **Try alternative terms**: If "deploy" doesn't work, try "deployment" or "ci-cd"
115
+ 3. **Check popular sources**: Many skills come from `vercel-labs/agent-skills` or `ComposioHQ/awesome-claude-skills`
116
+
117
+ ## When No Skills Are Found
118
+
119
+ If no relevant skills exist:
120
+
121
+ 1. Acknowledge that no existing skill was found
122
+ 2. Offer to help with the task directly using your general capabilities
123
+ 3. Suggest the user could create their own skill with `npx skills init`
124
+
125
+ Example:
126
+
127
+ ```
128
+ I searched for skills related to "xyz" but didn't find any matches.
129
+ I can still help you with this task directly! Would you like me to proceed?
130
+
131
+ If this is something you do often, you could create your own skill:
132
+ npx skills init my-xyz-skill
133
+ ```
@@ -0,0 +1 @@
1
+ 0.7.16