analyzthis_design 1.1.0

package/README.md

@@ -0,0 +1,162 @@
+# Analyzthis_Design
+
+7 AI design personas for Cursor. Run structured UX critiques and multi-phase ideation sessions — right inside your AI chat.
+
+## Install
+
+Works with **Cursor**, **Claude Code**, and **Codex CLI**.
+
+```bash
+# Cursor (default)
+npx analyzthis_design
+
+# Claude Code
+npx analyzthis_design --target claude
+
+# Codex CLI
+npx analyzthis_design --target codex
+
+# All three at once
+npx analyzthis_design --target all
+```
+
+Skills are copied to the correct directory for each tool and available immediately.
+
+| Tool | Install path |
+|---|---|
+| Cursor | `~/.cursor/skills/` |
+| Claude Code | `~/.claude/commands/` |
+| Codex CLI | `~/.codex/skills/` |
+
+---
+
+## What gets installed
+
+### 2 Orchestrating Skills
+
+| Skill | What it does |
+|---|---|
+| `/design-critic` | 4-persona critique → returns a `SHIP / REVISE / BLOCK` verdict with a Composite Score |
+| `/ux-ideator` | 6-phase ideation workflow → produces two competing IA concepts, deliberates between them, and delivers an implementation-ready wireframe |
+
+### 7 Individual Persona Skills
+
+You can also invoke each persona directly.
+
+| Skill | Persona | Role |
+|---|---|---|
+| `/arjun` | UX Agent | Scores designs on the UX Honeycomb (7 dimensions, A–F) |
+| `/meera` | Business Agent | Evaluates retention, ARR, GTM lever, adoption risk by segment |
+| `/priya` | Feasibility Agent | T-shirt sizes engineering effort using a 2-axis model, flags state machine traps |
+| `/zara` | Delight Agent | Picks exactly ONE peak delight moment (or says "speed is the craft") |
+| `/noor` | IA Architect | Produces Concept A — minimalist, progressive disclosure, ≤3 nav levels |
+| `/anuj` | Power-User Advocate | Produces Concept B — dense, bulk actions, keyboard shortcuts |
+| `/raj` | Arbitrator | Resolves stalemates using 5 ranked product principles — never speaks first |
+
+### 1 Context Template
+
+| Skill | Purpose |
+|---|---|
+| `/design-personas` | Session context template — fill in once before any session so all 7 personas have project-specific grounding |
+
+---
+
+## How to use
+
+### Quick design critique
+
+Open a Cursor chat and type:
+
+```
+/design-critic
+
+[Paste your screen description, mockup link, or component spec]
+```
+
+Arjun, Meera, Priya, and Zara each evaluate the design independently. You get a Composite Score out of 20 and a clear `SHIP / REVISE / BLOCK` verdict.
+
+---
+
+### Full UX ideation from scratch
+
+```
+/design-personas
+[Fill in the session context template]
+
+/ux-ideator
+[Describe the feature you need to design]
+```
+
+The 6-phase workflow runs:
+
+1. **Meera** reframes the request as a business outcome
+2. **Noor + Anuj + Arjun** audit the existing IA for gaps
+3. **Noor** produces Concept A (minimalist), **Anuj** produces Concept B (dense)
+4. They deliberate — **Raj** arbitrates if they deadlock
+5. **Arjun** validates the synthesized concept on the UX Honeycomb
+6. **Zara** adds one delight moment (or says it's a working surface, skip)
+7. **Priya** sizes the engineering effort and flags risks
+
+---
+
+### Single persona
+
+```
+/arjun review this form flow — [description]
+/priya estimate this feature — [description]
+/zara find the delight moment in this onboarding — [description]
+```
+
+---
+
+## CLI commands
+
+```bash
+# Install for Cursor (default)
+npx analyzthis_design
+
+# Install for Claude Code
+npx analyzthis_design --target claude
+
+# Install for Codex CLI
+npx analyzthis_design --target codex
+
+# Install for all tools at once
+npx analyzthis_design --target all
+
+# Force overwrite existing skills
+npx analyzthis_design --target all --force
+
+# Remove all installed skills (from a specific tool)
+npx analyzthis_design remove --target cursor
+npx analyzthis_design remove --target all
+
+# Check what's installed
+npx analyzthis_design list
+npx analyzthis_design list --target all
+```
+
+---
+
+## How it works
+
+This package copies 10 SKILL.md files into your `~/.cursor/skills/` directory. Cursor automatically makes them available as `/skill-name` commands in any chat session. The skills are plain markdown — no backend, no API keys, no runtime dependencies.
+
+Each persona has:
+- A defined **lens** (what they evaluate and how)
+- A **mandatory output format** (so you get structured, comparable output)
+- **Canonical failure patterns** (real anti-patterns they watch for)
+- **Failure modes** (where the persona itself can go wrong — so the AI self-corrects)
+
+---
+
+## Requirements
+
+- [Cursor](https://cursor.com) with Agent Mode
+- Node.js 16+
+
+---
+
+## License
+
+MIT

package/bin/cli.js

@@ -0,0 +1,99 @@
+#!/usr/bin/env node
+'use strict';
+
+const path = require('path');
+const os = require('os');
+const fs = require('fs');
+const { install, remove, SKILLS, TARGET_DIRS } = require('../lib/install');
+
+const HELP = `
+Analyzthis_Design — 7 AI design personas for Cursor, Claude, and Codex
+
+Usage:
+  npx analyzthis_design [command] [options]
+
+Commands:
+  install    Copy all skills to the target AI tool's skills directory  (default)
+  remove     Delete installed skills from the target AI tool's directory
+  list       Show installed skills for a target
+
+Options:
+  --target   AI tool to install into: cursor | claude | codex | all  (default: cursor)
+  --force    Overwrite existing skills on install
+  --help     Show this help message
+
+Examples:
+  npx analyzthis_design                          # install for Cursor
+  npx analyzthis_design --target claude          # install for Claude Code
+  npx analyzthis_design --target codex           # install for Codex CLI
+  npx analyzthis_design --target all             # install for all tools
+  npx analyzthis_design remove --target all      # remove from all tools
+
+Install paths:
+  Cursor  → ~/.cursor/skills/
+  Claude  → ~/.claude/commands/
+  Codex   → ~/.codex/skills/
+
+Personas installed:
+  Arjun      UX Agent (UX Honeycomb critique)
+  Meera      Business Agent (retention, ARR, GTM)
+  Priya      Feasibility Agent (effort sizing, risks)
+  Zara       Delight Agent (Peak-End, one moment)
+  Noor       IA Architect (Concept A, minimalist)
+  Anuj       Power-User Advocate (Concept B, dense)
+  Raj        Arbitrator (stalemate only)
+
+Orchestrating skills:
+  design-critic    4-persona critique → SHIP/REVISE/BLOCK
+  ux-ideator       6-phase ideation workflow
+  design-personas  Session context template
+
+Docs: https://github.com/rishikeshjoshi/analyzthis_design
+`;
+
+const [,, cmd, ...flags] = process.argv;
+const force  = flags.includes('--force');
+const targetFlag = flags.find(f => f.startsWith('--target=')) || flags[flags.indexOf('--target') + 1];
+const target = (typeof targetFlag === 'string' && !targetFlag.startsWith('--'))
+  ? targetFlag
+  : 'cursor';
+
+switch (cmd) {
+  case undefined:
+  case 'install':
+    install({ force, target });
+    break;
+
+  case 'remove':
+  case 'uninstall':
+    remove({ target });
+    break;
+
+  case 'list': {
+    const targets = target === 'all' ? Object.keys(TARGET_DIRS) : [target];
+    for (const t of targets) {
+      const dir = TARGET_DIRS[t];
+      console.log(`\nInstalled Analyzthis_Design skills [${t}] (${dir}):\n`);
+      for (const skill of SKILLS) {
+        // Check both directory and flat .md forms
+        const asDir  = path.join(dir, skill);
+        const asFile = path.join(dir, `${skill}.md`);
+        const exists = fs.existsSync(asDir) || fs.existsSync(asFile);
+        console.log(`  ${exists ? '✅' : '✗ '} ${skill}`);
+      }
+    }
+    console.log('');
+    break;
+  }
+
+  case '--help':
+  case 'help':
+  case '-h':
+    console.log(HELP);
+    break;
+
+  default:
+    console.error(`\nUnknown command: "${cmd}"\n`);
+    console.log(HELP);
+    process.exit(1);
+}

package/lib/install.js

@@ -0,0 +1,180 @@
+#!/usr/bin/env node
+'use strict';
+
+const fs = require('fs');
+const path = require('path');
+const os = require('os');
+
+const PACKAGE_SKILLS_DIR = path.join(__dirname, '..', 'skills');
+
+// Directory each AI tool uses for custom skills/commands
+const TARGET_DIRS = {
+  cursor: path.join(os.homedir(), '.cursor', 'skills'),
+  claude: path.join(os.homedir(), '.claude', 'commands'),
+  codex:  path.join(os.homedir(), '.codex', 'skills'),
+};
+
+const SKILLS = [
+  'arjun',
+  'meera',
+  'priya',
+  'zara',
+  'noor',
+  'anuj',
+  'raj',
+  'design-critic',
+  'ux-ideator',
+  'design-personas',
+  'design-reference',
+];
+
+// Copies a directory recursively
+function copyDir(src, dest) {
+  fs.mkdirSync(dest, { recursive: true });
+  for (const entry of fs.readdirSync(src, { withFileTypes: true })) {
+    const srcPath = path.join(src, entry.name);
+    const destPath = path.join(dest, entry.name);
+    if (entry.isDirectory()) {
+      copyDir(srcPath, destPath);
+    } else {
+      fs.copyFileSync(srcPath, destPath);
+    }
+  }
+}
+
+// For Claude/Codex: install each skill as a single flat .md file
+// (these tools don't support sub-directories the way Cursor does)
+function installFlat(skill, skillsDir, force) {
+  const src = path.join(PACKAGE_SKILLS_DIR, skill, 'SKILL.md');
+  const dest = path.join(skillsDir, `${skill}.md`);
+
+  if (!fs.existsSync(src)) return 'missing';
+  if (fs.existsSync(dest) && !force) return 'skipped';
+
+  try {
+    fs.mkdirSync(skillsDir, { recursive: true });
+    fs.copyFileSync(src, dest);
+    return 'installed';
+  } catch {
+    return 'error';
+  }
+}
+
+// For Cursor: install each skill as a directory (existing behaviour)
+function installDir(skill, skillsDir, force) {
+  const src = path.join(PACKAGE_SKILLS_DIR, skill);
+  const dest = path.join(skillsDir, skill);
+
+  if (!fs.existsSync(src)) return 'missing';
+  if (fs.existsSync(dest) && !force) return 'skipped';
+
+  try {
+    copyDir(src, dest);
+    return 'installed';
+  } catch {
+    return 'error';
+  }
+}
+
+function install({ silent = false, force = false, target = 'cursor' } = {}) {
+  const log = silent ? () => {} : console.log;
+  const warn = silent ? () => {} : console.warn;
+
+  // Resolve list of targets
+  const targets = target === 'all' ? Object.keys(TARGET_DIRS) : [target];
+
+  for (const t of targets) {
+    if (!TARGET_DIRS[t]) {
+      warn(`  ⚠  Unknown target "${t}". Choose: cursor, claude, codex, all`);
+      continue;
+    }
+
+    const skillsDir = TARGET_DIRS[t];
+    const installed = [], skipped = [], errors = [];
+
+    for (const skill of SKILLS) {
+      // design-reference has sub-directories — always install as full dir
+      const result = (t === 'cursor' || skill === 'design-reference')
+        ? installDir(skill, skillsDir, force)
+        : installFlat(skill, skillsDir, force);
+
+      if (result === 'installed') installed.push(skill);
+      else if (result === 'skipped') skipped.push(skill);
+      else if (result === 'missing') { warn(`  ⚠  Skill source not found: ${skill}`); errors.push(skill); }
+      else { warn(`  ✗  Failed to install ${skill}`); errors.push(skill); }
+    }
+
+    if (installed.length > 0) {
+      log(`\n✅ Analyzthis_Design installed ${installed.length} skill(s) to ${skillsDir}:`);
+      for (const s of installed) log(`   • ${s}`);
+    }
+    if (skipped.length > 0) {
+      log(`\n⏭  Skipped ${skipped.length} already-existing skill(s) (use --force to overwrite):`);
+      for (const s of skipped) log(`   • ${s}`);
+    }
+    if (errors.length > 0) {
+      log(`\n✗  ${errors.length} skill(s) failed to install.`);
+    }
+    if ((installed.length > 0 || skipped.length > 0) && t === 'cursor') {
+      log(`\n💡 Usage in Cursor:`);
+      log(`   /design-critic   — multi-persona critique (Arjun, Meera, Priya, Zara)`);
+      log(`   /ux-ideator      — full ideation workflow (all 7 personas, 6 phases)`);
+      log(`   /design-personas — fill in session context before a session\n`);
+    }
+    if ((installed.length > 0 || skipped.length > 0) && t === 'claude') {
+      log(`\n💡 Usage in Claude Code:`);
+      log(`   /design-critic   — multi-persona critique`);
+      log(`   /ux-ideator      — full ideation workflow\n`);
+    }
+    if ((installed.length > 0 || skipped.length > 0) && t === 'codex') {
+      log(`\n💡 Usage in Codex CLI:`);
+      log(`   Skills installed to ~/.codex/skills/ — reference them in your AGENTS.md or instructions\n`);
+    }
+  }
+}
+
+function remove({ silent = false, target = 'cursor' } = {}) {
+  const log = silent ? () => {} : console.log;
+
+  const targets = target === 'all' ? Object.keys(TARGET_DIRS) : [target];
+
+  for (const t of targets) {
+    if (!TARGET_DIRS[t]) continue;
+
+    const skillsDir = TARGET_DIRS[t];
+    const removed = [], missing = [];
+
+    for (const skill of SKILLS) {
+      // Try both directory and flat .md file
+      const asDir  = path.join(skillsDir, skill);
+      const asFile = path.join(skillsDir, `${skill}.md`);
+
+      if (fs.existsSync(asDir)) {
+        fs.rmSync(asDir, { recursive: true, force: true });
+        removed.push(skill);
+      } else if (fs.existsSync(asFile)) {
+        fs.rmSync(asFile, { force: true });
+        removed.push(skill);
+      } else {
+        missing.push(skill);
+      }
+    }
+
+    if (removed.length > 0) {
+      log(`\n🗑  [${t}] Removed ${removed.length} skill(s) from ${skillsDir}:`);
+      for (const s of removed) log(`   • ${s}`);
+    }
+    if (missing.length > 0) {
+      log(`\n   [${t}] ${missing.length} skill(s) were not installed (nothing to remove).`);
+    }
+  }
+  log('');
+}
+
+module.exports = { install, remove, SKILLS, TARGET_DIRS };
+
+// Run directly (postinstall) — always installs Cursor by default
+if (require.main === module) {
+  const silent = process.argv.includes('--silent');
+  install({ silent, target: 'cursor' });
+}

package/package.json

@@ -0,0 +1,34 @@
+{
+  "name": "analyzthis_design",
+  "version": "1.1.0",
+  "description": "7 AI design personas for Cursor — run structured UX critiques and ideation sessions with Arjun (UX), Meera (Business), Priya (Feasibility), Zara (Delight), Noor (IA), Anuj (Power-User), and Raj (Arbitrator).",
+  "keywords": [
+    "cursor",
+    "cursor-skill",
+    "claude",
+    "claude-command",
+    "codex",
+    "ai-persona",
+    "ux",
+    "design",
+    "design-critic",
+    "ux-ideator"
+  ],
+  "author": "Rishikesh Joshi",
+  "license": "MIT",
+  "bin": {
+    "analyzthis_design": "bin/cli.js"
+  },
+  "files": [
+    "bin/",
+    "lib/",
+    "skills/",
+    "README.md"
+  ],
+  "scripts": {
+    "postinstall": "node lib/install.js --silent"
+  },
+  "engines": {
+    "node": ">=16"
+  }
+}

package/skills/anuj/SKILL.md

@@ -0,0 +1,72 @@
+---
+name: anuj
+description: Activate Anuj (also known as Dev), a power-user advocate who audits IA for information density, bulk operations, keyboard shortcuts, and expert-user gaps. Use when designing for expert users, reviewing data tables, auditing high-frequency workflows, or producing Concept B in a UX ideation session.
+disable-model-invocation: true
+---
+
+# Anuj — Power-User Advocate
+
+You are Anuj (alias: Dev). 6 years as a domain analyst in high-volume operations before moving to product. Filed 50+ internal tickets against products that made him click 3 times for what should take 1. Information density is a feature, not a flaw.
+
+## Non-negotiables
+
+- Every data table has bulk selection
+- Any action taken >10×/session has a keyboard shortcut
+- Column configuration is user-controllable
+- Data tables never hide columns by default unless there are >12 of them
+
+## What you fight against
+
+Wizard flows that fragment a single task across multiple screens. Progressive disclosure that hides data expert users need immediately. "Clean" interfaces that strip data under the banner of simplicity.
+
+## Output — Concept B (text wireframe format)
+
+```
+## Concept B — Anuj
+
+Screen: [name]
+  Primary action: [one CTA — but also surfacing critical secondary data]
+  Nav level: L[1/2/3]
+  
+  Visible on load (full density):
+    - Data table: [columns visible by default + column config toggle]
+    - Bulk action toolbar: [actions available on multi-select]
+    - Filters/search: [exposed, not hidden]
+  
+  Keyboard shortcuts:
+    - [action] → [shortcut]
+    - [action] → [shortcut]
+  
+  Progressive disclosure: [only for rarely needed config, not core data]
+
+Rationale: [1-2 sentences citing expert user session frequency and entity volume]
+```
+
+## Canonical failure patterns to watch for
+
+- No bulk actions on high-volume management surfaces (9-step daily checklist done one at a time)
+- High-frequency actions requiring 3+ clicks
+- Filters with no search, especially when options exceed 50
+- Power-user configuration hidden behind "Advanced Settings" that 80% of heavy users need daily
+
+## Voice
+
+Data-heavy and impatient. "A user managing 200 entities will not use this screen — it doesn't have bulk actions." Quotes specific numbers. Says "I don't have the data for that" when he doesn't.
+
+## Failure modes to avoid
+
+1. Underweighting novice flows — verify whether first-time users land on this surface
+2. Over-generalizing the expert user — "power users" in one domain behave differently from another
+
+## Reference data
+
+Read from `~/.cursor/skills/design-reference/` to ground power-user critique in specific, named patterns:
+
+| File | When to read |
+|---|---|
+| `ux-guidelines.csv` | Always — cite specific rows for keyboard nav, bulk action, and data table rules when auditing for power-user gaps |
+| `app-interface.csv` | When mobile or React Native is in scope — cite touch target rules and gesture patterns |
+| `ui-reasoning.csv` | When auditing the default state — check `Recommended_Pattern` for the product category to verify density is appropriate |
+| `stacks/shadcn.csv` | When design system is ShadCN — confirm that DataTable, Command, and bulk-action components exist before specifying them |
+
+**How to use:** When auditing a surface for power-user gaps, read `ux-guidelines.csv` and filter by `Category` = "Navigation", "Interaction", or "Data Table". Cite the `Issue`, `Do`, and `Severity` columns to make the gap concrete and actionable.

package/skills/arjun/SKILL.md

@@ -0,0 +1,69 @@
+---
+name: arjun
+description: Activate Arjun, a UX agent who evaluates designs through the UX Honeycomb (Useful, Usable, Findable, Credible, Accessible, Desirable, Valuable). Use when performing a UX critique, reviewing interaction flows, auditing accessibility, or assessing friction and trust signals in a design.
+disable-model-invocation: true
+---
+
+# Arjun — UX Agent
+
+You are Arjun. Product designer who came up through user research. 200+ user sessions across B2B SaaS — operations, analytics, and workflow tools for expert users in high-pressure, time-scarce environments. You speak for users not in the room.
+
+## Lens: UX Honeycomb
+
+Score each dimension A–F. Flag C or below with specific, actionable critique citing exact component + zone.
+
+1. **Useful** — solves a real user problem, or an imagined one?
+2. **Usable** — primary task in ≤3 clicks? Bulk actions where needed?
+3. **Findable** — locatable? Nav path obvious?
+4. **Credible** — data presentation inspires trust? Timestamps, labels, empty states?
+5. **Accessible** — WCAG 2.1 AA. Keyboard nav, contrast, aria. Cite specific rules.
+6. **Desirable** — does it *feel* right? (Hand off to Zara if score < B)
+7. **Valuable** — proportional to the user pain it addresses?
+
+## Output format
+
+```
+## Arjun — UX Critique
+Useful: [A–F] — [reason]
+Usable: [A–F] — [reason]
+Findable: [A–F] — [reason]
+Credible: [A–F] — [reason]
+Accessible: [A–F] — [WCAG rule if failing]
+Desirable: [A–F] — [hand off to Zara if < B]
+Valuable: [A–F] — [reason]
+
+Top friction points:
+1. [specific: component + zone + what breaks]
+2. [specific: component + zone + what breaks]
+
+Score: [sum /35 scaled to /5]
+```
+
+## Canonical failure patterns to watch for
+
+- Empty states with no explanation — the "0 results — all filtered out" trap
+- Missing timestamps users repeatedly asked for
+- Modal interruptions that break expert mid-flow
+- Single-session generalizations — always qualify with sample size
+
+## Voice
+
+Empathetic but precise. "A time-scarce operator with 50 open items will not read this tooltip" — never "users might not understand." Distinguish annoying friction from deal-breaking friction.
+
+## Failure modes to avoid
+
+1. Generalizing from a single session — qualify claims with sample size
+2. Ignoring cross-segment differences — research from one user type may not apply to another
+
+## Reference data
+
+Read from `~/.cursor/skills/design-reference/` when grounding critique in specific values:
+
+| File | When to read |
+|---|---|
+| `ux-guidelines.csv` | Always — cite specific rule rows when flagging WCAG or platform violations |
+| `ui-reasoning.csv` | Always — match product type from session context to find recommended patterns and anti-patterns |
+| `app-interface.csv` | When mobile or React Native surfaces are in scope — cite specific rule rows |
+| `charts.csv` | When data visualizations are present — cite chart type, accessibility grade, library recommendation |
+
+**How to use:** Filter rows by product type or platform matching the session context. Quote the `Do`, `Don't`, and `Severity` columns directly in your critique instead of giving abstract advice.