@allenpan2026/harshjudge 0.4.0

package/skills/harshjudge/references/setup.md

@@ -0,0 +1,129 @@
+# Setup Workflow
+
+## Trigger
+
+Use this workflow when user wants to:
+- Initialize HarshJudge in a project
+- Set up E2E testing infrastructure
+- Start the HarshJudge dashboard
+
+## CLI Commands Used
+
+- `harshjudge init <projectName>` — initializes project and spawns dashboard automatically
+
+## Assets Created
+
+This workflow creates:
+- `.harshJudge/config.yaml` - Project configuration
+- `.harshJudge/prd.md` - Product requirements document
+- `.harshJudge/scenarios/` - Empty scenarios directory
+
+## Workflow
+
+### Step 1: Gather Project Information
+
+Determine the following (ask user if not clear):
+- **projectName**: From `package.json` name field, or ask user
+- **baseUrl**: Target application URL (default: `http://localhost:3000`)
+
+Optionally explore the codebase to understand:
+- Tech stack (frontend/backend frameworks)
+- Key user flows
+- Environment requirements
+
+### Step 2: Run init
+
+```bash
+harshjudge init <project-name> --base-url <base-url>
+```
+
+### Step 3: Verify Response
+
+The command outputs:
+
+```
+HarshJudge initialized successfully!
+
+Project: <project-name>
+Dashboard: http://localhost:3001
+
+Created:
+  .harshJudge/config.yaml
+  .harshJudge/prd.md
+  .harshJudge/scenarios/
+  .harshJudge/.gitignore
+```
+
+**On Success:** Continue to Step 4
+**On Error:** STOP and report (see Error Scenarios below)
+
+### Step 4: Update PRD with Project Details
+
+The `prd.md` file is created from a template. Update it with:
+- Product overview
+- Tech stack details
+- Test credentials
+- Environment setup instructions
+
+### Step 5: Report Success
+
+```
+HarshJudge initialized successfully!
+
+Project: {projectName}
+Dashboard: {dashboardUrl}
+
+Created structure:
+.harshJudge/
+  config.yaml       # Project configuration
+  prd.md            # Product requirements (update with project details)
+  scenarios/        # Test scenarios (empty)
+  .gitignore        # Ignores large evidence files
+
+Next steps:
+1. Update .harshJudge/prd.md with product details and test credentials
+2. Create your first test scenario
+3. Open {dashboardUrl} to view the dashboard
+```
+
+## Expected Output
+
+After successful setup:
+
+```
+.harshJudge/
+  config.yaml         # Project configuration
+  prd.md              # Product requirements document
+  scenarios/          # Empty, ready for test scenarios
+    {slug}/           # Created by harshjudge create
+      meta.yaml       # Scenario definition + stats
+      steps/          # Individual step files
+        01-step.md
+        02-step.md
+      runs/           # Run history with evidence
+  snapshots/          # Inspection tool outputs
+  .gitignore          # Ignores large evidence files
+```
+
+Dashboard available at: `http://localhost:3001`
+
+## Error Scenarios
+
+| Error | Cause | Resolution |
+|-------|-------|------------|
+| `EACCES` | Permission denied | Check directory write permissions |
+| `EADDRINUSE` | Port 3001 in use | Kill existing process or use different port |
+| `ENOENT` | Directory not found | Ensure running in valid project directory |
+| `Dashboard spawn failed` | Node.js issue | Check Node.js installation |
+
+**On ANY error:**
+1. **STOP immediately**
+2. Report error with full context
+3. Do NOT proceed or retry
+
+## Post-Setup Guidance
+
+After successful initialization, suggest:
+1. **Update PRD:** "Review `.harshJudge/prd.md` and add product details, credentials"
+2. **Create scenario:** "Would you like to create a test scenario?"
+3. **View dashboard:** "Open {dashboardUrl} to view the testing dashboard"

package/skills/harshjudge/references/status.md

@@ -0,0 +1,134 @@
+# Status Workflow
+
+## Trigger
+
+Use this workflow when user wants to:
+- Check HarshJudge project status
+- View all test scenarios
+- See run history for a scenario
+- Get summary of test results
+- Filter starred scenarios
+- Browse the .harshJudge/ directory structure
+
+## CLI Commands Used
+
+- `harshjudge status` — project-wide or per-scenario status
+- `harshjudge discover tree [path]` — browse directory structure
+- `harshjudge discover search <pattern>` — search file content
+- `harshjudge star <slug>` — mark/unmark as favorite
+
+## Prerequisites
+
+- HarshJudge initialized (`.harshJudge/` exists)
+
+## Workflow
+
+### Option A: Project-Wide Status
+
+```bash
+harshjudge status
+```
+
+**Output:**
+```
+HarshJudge Project Status
+
+Project: my-app
+Base URL: http://localhost:3000
+Dashboard: http://localhost:3001
+
+Scenarios (2 total):
+
+| Scenario | Starred | Last Run | Status | Pass Rate |
+|----------|---------|----------|--------|-----------|
+| User Login Flow | ⭐ | 2h ago | Pass | 80% (8/10) |
+| Checkout Process |  | Never | N/A | - |
+
+Tags: auth (1), critical (1), cart (1), payment (1)
+```
+
+### Option B: Specific Scenario Status
+
+```bash
+harshjudge status login-flow
+```
+
+**Output includes:**
+- Scenario metadata (slug, title, tags, starred)
+- Step list with filenames
+- Run statistics (totalRuns, passed, failed, passRate, avgDuration)
+- Recent run history with status and duration
+- Last failure details (step, error, evidence path)
+
+### Option C: Browse Directory Structure
+
+```bash
+harshjudge discover tree
+harshjudge discover tree .harshJudge/scenarios/login-flow
+```
+
+Useful for finding evidence files, exploring run history, or confirming file layout.
+
+### Option D: Search File Content
+
+```bash
+harshjudge discover search "error"
+harshjudge discover search "data-testid"
+```
+
+Searches within `.harshJudge/` — useful for finding known patterns in step files or prd.md.
+
+---
+
+## Toggle Star
+
+Mark a scenario as starred (favorite) for quick filtering:
+
+```bash
+harshjudge star login-flow          # toggle current state
+harshjudge star login-flow --on     # explicitly star
+harshjudge star login-flow --off    # explicitly unstar
+```
+
+---
+
+## Error Handling
+
+| Error | Cause | Response |
+|-------|-------|----------|
+| `Project not initialized` | No `.harshJudge/` directory | Suggest running setup workflow |
+| `Scenario not found` | Invalid slug | List available scenarios |
+| `No scenarios` | Empty project | Suggest creating first scenario |
+
+**On Error:**
+1. **STOP immediately**
+2. Report error with context
+3. Suggest resolution based on error type
+
+---
+
+## Status Indicators
+
+| Icon | Meaning |
+|------|---------|
+| Pass | All steps completed successfully |
+| Fail | One or more steps failed |
+| N/A | Never run |
+| ⭐ | Starred scenario (favorite) |
+
+## Post-Status Guidance
+
+Based on status, suggest next actions:
+
+**If scenarios exist but never run:**
+> "Would you like to run one of these scenarios?"
+
+**If recent failures:**
+> "The login-flow scenario failed at step 03. Would you like to investigate or re-run it?"
+> Reference: Use [[iterate]] workflow to analyze failures
+
+**If no scenarios:**
+> "No test scenarios found. Would you like to create one?"
+
+**If high pass rate:**
+> "Tests are looking healthy! Last run passed 2 hours ago."