prjct-cli 1.22.0 → 1.24.0

package/templates/commands/review.md

@@ -1,276 +0,0 @@
----
-allowed-tools: [Bash, Read, Write, Task, AskUserQuestion]
-description: 'Code review with MCP agent and GitHub approvals'
-timestamp-rule: 'GetTimestamp() for ALL timestamps'
-architecture: 'Write-Through (JSON -> MD -> Events)'
-storage-layer: true
-source-of-truth: 'storage/state.json'
----
-
-# /p:review
-
-Run MCP code review agent and wait for GitHub PR approvals.
-
-## Usage
-
-```
-/p:review [--skip-mcp]    # Skip MCP agent review
-```
-
-## Context Variables
-- `{projectId}`: From `.prjct/prjct.config.json`
-- `{globalPath}`: `~/.prjct-cli/projects/{projectId}`
-- `{statePath}`: `{globalPath}/storage/state.json`
-- `{memoryPath}`: `{globalPath}/memory/events.jsonl`
-- `{syncPath}`: `{globalPath}/sync/pending.json`
-
-## Step 1: Validate Project
-
-READ: `.prjct/prjct.config.json`
-EXTRACT: `projectId`
-
-IF file not found:
-  OUTPUT: "No prjct project. Run /p:init first."
-  STOP
-
-## Step 2: Validate Workflow Phase
-
-READ: `{globalPath}/storage/state.json`
-
-IF currentTask is null:
-  OUTPUT: "No active task. Use p. task to start one."
-  STOP
-
-IF currentTask.workflow exists:
-  IF currentTask.workflow.phase != "test":
-    OUTPUT:
-    ```
-    Cannot start review. Current phase: {currentTask.workflow.phase}
-
-    Required phase: test
-
-    Workflow: analyze → branch → implement → test → review → merge → ship → verify
-    
-    Run p. test first to advance to test phase.
-    ```
-    STOP
-
-## Step 3: Run MCP Code Review (unless --skip-mcp)
-
-IF NOT --skip-mcp:
-  OUTPUT: "Running MCP code review..."
-  
-  ### Get changed files
-  BASH: `git diff --name-only HEAD~1..HEAD 2>/dev/null || git diff --name-only`
-  SET: {changedFiles} = result
-  
-  ### Analyze with MCP agent
-  FOR each file in {changedFiles}:
-    READ: file
-    ANALYZE for:
-    - Security issues (hardcoded secrets, injection vulnerabilities)
-    - Logic errors
-    - Missing error handling
-    - Performance issues
-    - Code style violations
-    
-    ASSIGN confidence score (0-100):
-    - 90-100: Definite bug/security issue
-    - 70-89: Likely problem
-    - 50-69: Maybe a problem
-    - 0-49: Nitpick/style
-  
-  SET: {issues} = issues with confidence >= 70
-  SET: {mcpScore} = 100 - (count of high-confidence issues * 10)
-  
-  IF {issues}.length > 0:
-    OUTPUT:
-    ```
-    ## MCP Code Review Results
-    
-    Found {issues.length} issues (confidence >= 70%):
-    
-    {FOR each issue:}
-    - [{confidence}%] {description}
-      File: {file}:{line}
-    {END FOR}
-    ```
-    
-    USE AskUserQuestion:
-    ```
-    question: "Code review found {issues.length} issues. How to proceed?"
-    header: "Review Issues"
-    options:
-      - label: "Fix issues first"
-        description: "Return to implement phase to fix"
-      - label: "Proceed anyway"
-        description: "Continue with PR creation"
-    ```
-    
-    IF choice == "Fix issues first":
-      OUTPUT: "Returning to implement phase. Fix issues and run p. test again."
-      STOP
-  ELSE:
-    OUTPUT: "✓ MCP review passed. No high-confidence issues found."
-    SET: {mcpScore} = 100
-
-## Step 4: Create/Check PR
-
-### Check if PR exists
-BASH: `gh pr view --json url,number,state 2>/dev/null`
-
-IF PR exists:
-  SET: {prUrl} = result.url
-  SET: {prNumber} = result.number
-  SET: {prState} = result.state
-  OUTPUT: "PR exists: {prUrl}"
-ELSE:
-  OUTPUT: "Creating PR..."
-  
-  SET: {branchName} = currentTask.branch.name
-  SET: {baseBranch} = currentTask.branch.baseBranch OR "main"
-  
-  BASH: `git push -u origin {branchName} 2>&1`
-  
-  SET: {prTitle} = "{currentTask.type}: {currentTask.description}"
-  SET: {prBody} = """
-## Summary
-{currentTask.description}
-
-## Workflow Phase
-- [x] Analyze
-- [x] Branch
-- [x] Implement  
-- [x] Test
-- [ ] Review ← current
-- [ ] Merge
-- [ ] Ship
-- [ ] Verify
-
-## MCP Review Score
-{mcpScore}/100
-
----
-Generated with [p/](https://www.prjct.app/)
-"""
-  
-  BASH: `gh pr create --title "{prTitle}" --base {baseBranch} --body "$(cat <<'PREOF'
-{prBody}
-PREOF
-)"`
-  
-  EXTRACT: {prUrl}, {prNumber} from output
-
-## Step 5: Check GitHub Approvals
-
-OUTPUT: "Checking for approvals..."
-
-BASH: `gh pr view {prNumber} --json reviews,reviewDecision`
-SET: {reviews} = result.reviews
-SET: {decision} = result.reviewDecision
-
-IF {decision} == "APPROVED":
-  SET: {approved} = true
-  SET: {approvals} = reviews where state == "APPROVED"
-  OUTPUT: "✓ PR approved by {approvals.length} reviewer(s)"
-ELSE IF {decision} == "CHANGES_REQUESTED":
-  OUTPUT:
-  ```
-  ⚠️ Changes requested
-  
-  {FOR each review where state == "CHANGES_REQUESTED":}
-  - {review.author}: {review.body}
-  {END FOR}
-  
-  Address feedback, push changes, and run p. review again.
-  ```
-  STOP
-ELSE:
-  OUTPUT:
-  ```
-  ⏳ Waiting for approvals
-  
-  PR: {prUrl}
-  
-  Request review from team members, then run p. review again.
-  ```
-  STOP
-
-## Step 6: Update Workflow Phase
-
-SET: {now} = GetTimestamp()
-
-SET: currentTask.workflow.phase = "review"
-SET: currentTask.workflow.checkpoints.review = {
-  "completedAt": "{now}",
-  "data": {
-    "mcpScore": {mcpScore},
-    "approvals": {approvals},
-    "prUrl": "{prUrl}",
-    "prNumber": {prNumber}
-  }
-}
-SET: currentTask.workflow.lastCheckpoint = "review"
-SET: currentTask.branch.prUrl = "{prUrl}"
-SET: currentTask.branch.prNumber = {prNumber}
-
-WRITE: `{statePath}`
-
-## Step 7: Log Events
-
-APPEND to `{memoryPath}`:
-```json
-{"timestamp":"{now}","action":"phase_advanced","taskId":"{currentTask.id}","from":"test","to":"review"}
-{"timestamp":"{now}","action":"checkpoint_completed","taskId":"{currentTask.id}","checkpoint":"review","data":{"mcpScore":{mcpScore},"approvals":{approvals.length}}}
-```
-
-APPEND to `{syncPath}`:
-```json
-{"type":"workflow.phase_advanced","data":{"taskId":"{currentTask.id}","from":"test","to":"review","prUrl":"{prUrl}"},"timestamp":"{now}"}
-```
-
-## Output
-
-```
-✓ Review Complete
-
-Task: {currentTask.description}
-MCP Score: {mcpScore}/100
-Approvals: {approvals.length}
-PR: {prUrl}
-
-Phase: review (5/11 checkpoints)
-
-Workflow:
-1. analyze ✓
-2. branch ✓
-3. implement ✓
-4. test ✓
-5. review ✓
-6. merge ← next
-
-Next: p. merge to merge PR
-```
-
-## Error Handling
-
-| Error | Response | Action |
-|-------|----------|--------|
-| No project | "No prjct project" | STOP |
-| No active task | "No active task" | STOP |
-| Wrong phase | Show required phase | STOP |
-| MCP issues found | Ask user | WAIT |
-| Changes requested | Show feedback | STOP |
-| No approvals | Show PR URL | STOP |
-| gh CLI missing | "Install gh CLI" | STOP |
-
-## Natural Language Triggers
-
-- `p. review` -> /p:review
-- `p. code review` -> /p:review
-- `p. pr` -> /p:review
-
-## References
-
-- Architecture: `~/.prjct-cli/docs/architecture.md`
-- Workflow: `~/.prjct-cli/docs/workflow.md`

package/templates/commands/serve.md

@@ -1,118 +0,0 @@
----
-allowed-tools: [Read, Bash]
-description: 'Start prjct web server for dashboard access'
-timestamp-rule: 'GetTimestamp() for session start'
-architecture: 'HTTP server with REST API and SSE'
----
-
-# /p:serve - Start Web Server
-
-Starts the prjct HTTP server for web dashboard access and API.
-
-## Usage
-
-```
-/p:serve [port]
-```
-
-- `port`: Optional port number (default: 3478)
-
-## Flow
-
-### Step 1: Validate Project
-1. Read `.prjct/prjct.config.json` → extract projectId
-2. If not found: "No prjct project. Run /p:init first." → STOP
-
-### Step 2: Check Port
-1. Default port: 3478 ("prjct" on phone keypad)
-2. If port specified, validate it's a number between 1024-65535
-3. Check if port is available
-
-### Step 3: Start Server
-Execute with Bun:
-```bash
-bun -e "
-const { startServer } = require('./core/server');
-startServer('{projectId}', '{projectPath}', {port});
-"
-```
-
-### Step 4: Output Server Info
-
-```
-🚀 prjct server started
-
-   URL: http://localhost:{port}
-   Project: {projectId}
-
-   Endpoints:
-   - GET  /api/state      Current task
-   - GET  /api/queue      Task queue
-   - GET  /api/ideas      Ideas backlog
-   - GET  /api/roadmap    Feature roadmap
-   - GET  /api/shipped    Shipped items
-   - GET  /api/dashboard  Combined data
-   - GET  /api/events     SSE stream
-
-   Press Ctrl+C to stop
-```
-
-## API Endpoints
-
-| Endpoint | Method | Description |
-|----------|--------|-------------|
-| `/health` | GET | Server health check |
-| `/api/state` | GET | Current task state |
-| `/api/queue` | GET | Task queue |
-| `/api/ideas` | GET | Ideas backlog |
-| `/api/roadmap` | GET | Feature roadmap |
-| `/api/shipped` | GET | Shipped items |
-| `/api/dashboard` | GET | All data combined |
-| `/api/events` | GET | SSE real-time stream |
-| `/api/context/:name` | GET | Context markdown files |
-
-## Real-Time Updates (SSE)
-
-Connect to `/api/events` for live updates:
-
-```javascript
-const events = new EventSource('http://localhost:3478/api/events');
-
-events.addEventListener('task:started', (e) => {
-  console.log('Task started:', JSON.parse(e.data));
-});
-
-events.addEventListener('task:completed', (e) => {
-  console.log('Task completed:', JSON.parse(e.data));
-});
-```
-
-## Event Types
-
-- `connected` - Initial connection established
-- `heartbeat` - Keep-alive ping (every 30s)
-- `task:started` - New task started
-- `task:completed` - Task finished
-- `task:paused` - Task paused
-- `feature:shipped` - Feature shipped
-- `state:updated` - State changed
-- `queue:updated` - Queue changed
-
-## Error Handling
-
-| Error | Response |
-|-------|----------|
-| Project not initialized | Exit with message |
-| Port in use | Suggest alternative port |
-| Permission denied | Request elevated permissions |
-
-## Output Format
-
-```
-🚀 prjct server started
-
-   URL: http://localhost:3478
-   Project: abc-123-def
-
-   Dashboard ready at http://localhost:3478
-```

package/templates/commands/setup.md

@@ -1,91 +0,0 @@
-# /p:setup - Reconfigure prjct-cli Installation
-
-Reconfigures prjct-cli installation for Claude Code and Claude Desktop.
-
-## Usage
-
-```
-/p:setup [--force]
-```
-
-## What This Command Does
-
-1. **Detects Claude Installation**
-   - Checks if `~/.claude/` directory exists
-   - Verifies Claude Code or Claude Desktop is installed
-
-2. **Syncs Commands to Claude**
-   - Updates all `/p:*` commands in `~/.claude/commands/p/`
-   - Adds new commands from latest version
-   - Updates existing commands with latest templates
-   - Removes orphaned/deprecated commands
-
-3. **Installs MCP Servers**
-   - Reads `templates/mcp-config.json` for MCP server definitions
-   - Merges into `~/.claude/settings.json` (preserves existing settings)
-   - Installs Context7 for library documentation lookup
-   - Installs Atlassian for JIRA/Confluence (OAuth, SSO compatible)
-   - Does NOT overwrite existing MCP configurations
-
-4. **Installs/Updates Global Configuration**
-   - Creates or updates `~/.claude/CLAUDE.md`
-   - Adds prjct-specific instructions for Claude
-   - Adds Context7 and Atlassian usage instructions
-   - Preserves existing user configuration
-
-5. **Reports Results**
-   - Shows commands added, updated, removed
-   - Shows MCP servers installed
-   - Displays any errors encountered
-   - Confirms successful installation
-
-## Options
-
-- `--force`: Remove existing installation and reinstall from scratch
-
-## When to Use
-
-- **After updating prjct-cli**: `npm update -g prjct-cli && /p:setup`
-- **Commands not working**: If `/p:*` commands aren't recognized
-- **Fresh installation**: After installing on a new machine
-- **Troubleshooting**: When encountering command-related issues
-
-## Requirements
-
-- Claude Code or Claude Desktop must be installed
-- Write permissions to `~/.claude/` directory
-
-## Output Example
-
-```
-🔧 Reconfiguring prjct...
-
-📦 Installing /p:* commands...
-✓ 3 new, 12 updated, 1 removed
-
-🔌 Installing MCP servers...
-✓ context7 (library documentation)
-✓ Atlassian (JIRA/Confluence via OAuth)
-
-📝 Installing global configuration...
-✓ Updated ~/.claude/CLAUDE.md
-✓ Updated ~/.claude/settings.json
-
-✅ Setup complete!
-
-MCP Tools Available:
-• context7: resolve-library-id, get-library-docs
-• Atlassian: jira_search_issues, jira_get_issue, jira_transition_issue
-```
-
-## Error Handling
-
-- **Claude not detected**: Shows installation URLs for Claude Code/Desktop
-- **Permission errors**: Reports which files couldn't be written
-- **Template errors**: Lists which commands failed to install
-
-## Notes
-
-- This command does NOT require an initialized prjct project
-- Safe to run multiple times (idempotent)
-- Will not overwrite user customizations in `~/.claude/CLAUDE.md`