claude-dev-cli 0.16.1__tar.gz → 0.18.0__tar.gz

{claude_dev_cli-0.16.1/src/claude_dev_cli.egg-info → claude_dev_cli-0.18.0}/PKG-INFO

@@ -1,7 +1,7 @@
 Metadata-Version: 2.4
 Name: claude-dev-cli
-Version: 0.16.1
-Summary: A powerful CLI tool for developers using Claude AI with multi-API routing, test generation, code review, and usage tracking
+Version: 0.18.0
+Summary: A powerful CLI tool for developers using Claude AI with multi-API routing, test generation, code review, usage tracking, and project management automation
 Author-email: Julio <thinmanj@users.noreply.github.com>
 License: MIT
 Project-URL: Homepage, https://github.com/thinmanj/claude-dev-cli
@@ -47,6 +47,23 @@ Requires-Dist: requests>=2.28.0; extra == "local"
 Provides-Extra: all-providers
 Requires-Dist: openai>=1.0.0; extra == "all-providers"
 Requires-Dist: requests>=2.28.0; extra == "all-providers"
+Provides-Extra: project-mgmt
+Requires-Dist: repo-tickets>=0.8.0; extra == "project-mgmt"
+Requires-Dist: python-docx>=0.8.11; extra == "project-mgmt"
+Provides-Extra: notifications
+Requires-Dist: requests>=2.28.0; extra == "notifications"
+Requires-Dist: python-telegram-bot>=20.0; extra == "notifications"
+Provides-Extra: vcs
+Requires-Dist: GitPython>=3.1.0; extra == "vcs"
+Provides-Extra: environment
+Requires-Dist: docker>=6.0.0; extra == "environment"
+Provides-Extra: all-pm
+Requires-Dist: repo-tickets>=0.8.0; extra == "all-pm"
+Requires-Dist: python-docx>=0.8.11; extra == "all-pm"
+Requires-Dist: requests>=2.28.0; extra == "all-pm"
+Requires-Dist: python-telegram-bot>=20.0; extra == "all-pm"
+Requires-Dist: GitPython>=3.1.0; extra == "all-pm"
+Requires-Dist: docker>=6.0.0; extra == "all-pm"
 Provides-Extra: dev
 Requires-Dist: pytest>=7.0.0; extra == "dev"
 Requires-Dist: black>=23.0.0; extra == "dev"
@@ -68,10 +85,113 @@ Dynamic: license-file
 [![Homebrew](https://img.shields.io/badge/homebrew-available-orange.svg)](https://github.com/thinmanj/homebrew-tap)
 [![Code style: black](https://img.shields.io/badge/code%20style-black-000000.svg)](https://github.com/psf/black)
 
-A powerful command-line tool for developers using Claude AI with multi-API routing, test generation, code review, and comprehensive usage tracking.
+A powerful command-line tool for developers using Claude AI with multi-API routing, test generation, code review, comprehensive usage tracking, and **AI-powered project automation**.
 
 ## Features
 
+### 🧠 Intelligent Context Gathering (v0.18.0+) ✨ **NEW**
+
+**AI learns your codebase patterns before generating code:**
+
+```bash
+# Automatically gathers context (language, framework, conventions, similar code)
+cdc ticket execute TASK-123 --commit --notify
+
+# Skip context gathering for faster execution
+cdc ticket execute TASK-456 --no-context
+```
+
+**What context is gathered:**
+- 🗂️ Language & Framework (Python/Django, JS/React, Go, etc.)
+- 📦 Dependencies (requirements.txt, package.json, go.mod)
+- 📁 Directory Structure (models/, views/, controllers/, tests/)
+- 📝 Naming Conventions (snake_case, camelCase, PascalCase)
+- 🔍 Similar Existing Code (based on ticket keywords)
+- ⚙️ Common Imports & Patterns
+
+**Result:** AI generates code that follows YOUR project's patterns, not generic templates!
+
+See [CONTEXT_INTELLIGENCE.md](CONTEXT_INTELLIGENCE.md) for full details.
+
+### 🤖 AI-Powered Project Automation (v0.17.0+)
+
+**Transform tickets into working code automatically:**
+
+```bash
+# Execute a ticket end-to-end: fetch → analyze → generate code → tests → commit
+cdc ticket execute TASK-123 --commit --notify
+```
+
+**What happens:**
+1. 📥 Fetches ticket from backend (repo-tickets, Jira, Linear, GitHub Issues)
+2. 🧠 AI analyzes requirements and acceptance criteria
+3. 💻 Generates production-ready implementation code
+4. ✅ Generates comprehensive tests
+5. 📝 Commits to git with conventional format + co-author
+6. 🎫 Updates ticket status to "completed"
+7. 📊 Logs all progress to `.cdc-logs/progress.md`
+8. 🔔 Sends push notifications (FREE via ntfy.sh)
+
+**Ticket Management:**
+```bash
+# Create tickets
+cdc tickets create "Add user authentication" --priority high --type feature
+
+# List tickets
+cdc tickets list --status open
+
+# Multi-backend support
+cdc ticket execute TASK-456 --backend repo-tickets
+```
+
+**AI-Powered Bug Triage:**
+```bash
+# Report bugs with automatic severity/category classification
+cdc bug report \
+  --title "App crashes on startup" \
+  --description "Application fails to launch" \
+  --expected "App should start normally" \
+  --actual "App crashes immediately" \
+  --steps "Open app" --steps "Wait 2 seconds" \
+  --environment production
+
+# AI automatically assigns:
+# ✅ Severity: CRITICAL (crash detected)
+# ✅ Category: CRASH
+# ✅ Priority: critical
+# ✅ Sends urgent notification
+# ✅ Detects duplicates
+```
+
+**Progress Tracking:**
+```bash
+# Initialize logging
+cdc log init "My Project"
+
+# Add entries
+cdc log entry "Completed auth module" --ticket TASK-123 --level success
+
+# Generate reports
+cdc log report  # Beautiful markdown summary with stats
+```
+
+**Features:**
+- 🎫 **Ticket Backends**: repo-tickets, Jira, Linear, GitHub Issues, Markdown (fallback)
+- 🤖 **Full AI Automation**: Requirements → Code → Tests → Commit → Deploy
+- 🐛 **Smart Bug Triage**: AI classifies severity, category, priority
+- 📊 **Progress Logging**: Markdown-based tracking with artifact linking
+- 🔔 **Notifications**: FREE push notifications via ntfy.sh (no signup!)
+- 🔄 **VCS Integration**: Git commits with co-author attribution
+- 📝 **Structured Data**: BDD scenarios, acceptance criteria, stack traces
+- 🔍 **Duplicate Detection**: Find similar bugs automatically
+- ⚡ **Cost Optimization**: Use FREE local AI (Ollama) for code generation
+
+**Install with project management:**
+```bash
+pip install 'claude-dev-cli[all-pm]'  # All project management features
+pip install 'claude-dev-cli[project-mgmt]'  # Ticket management only
+```
+
 ### 🌐 Multi-Provider AI Support (v0.14.0+)
 - **Anthropic (Claude)**: GPT-4 class models with 200k context
   - Haiku, Sonnet, Opus - full model family
@@ -175,6 +295,17 @@ A powerful command-line tool for developers using Claude AI with multi-API routi
 - **Project Memory**: Remember preferences per project
 - **Global Config**: Set context defaults in `~/.claude-dev-cli/config.json`
 
+### 🔄 Workflows (v0.7.0+, Enhanced v0.16.1)
+- **Multi-Step Automation**: Chain AI operations and shell commands with YAML
+- **AI Decision Making**: Use `ask` command to make intelligent workflow decisions (v0.16.1)
+- **Variable Interpolation**: Pass data between steps with `{{variable}}` syntax
+- **Conditional Execution**: Skip steps based on conditions with `if` clauses
+- **Approval Gates**: Human-in-the-loop with `approval_required: true`
+- **Model Selection**: Use different AI models per step for cost optimization (v0.16.1)
+- **Error Handling**: Continue on failure with `continue_on_error: true`
+- **Three Step Types**: command (cdc), shell (bash/zsh), set (variables)
+- **Cost Optimization**: Mix free local AI with paid cloud AI strategically
+
 ### 🎒 TOON Format Support (Optional)
 - **Token Reduction**: 30-60% fewer tokens than JSON
 - **Cost Savings**: Reduce API costs significantly
@@ -251,6 +382,30 @@ pip install 'claude-dev-cli[toon]'
 Adds:
 - `toon-format>=0.1.0` - TOON encoding/decoding
 
+**Project Management** (v0.17.0+ - ticket automation, bug triage):
+
+```bash
+# Core project management (tickets, logging)
+pip install 'claude-dev-cli[project-mgmt]'
+
+# With notifications
+pip install 'claude-dev-cli[notifications]'
+
+# With VCS integration
+pip install 'claude-dev-cli[vcs]'
+
+# All project management features
+pip install 'claude-dev-cli[all-pm]'
+```
+
+Adds:
+- `repo-tickets>=0.8.0` - Ticket management backend
+- `python-docx>=0.8.11` - Document processing
+- `requests>=2.28.0` - HTTP client for notifications
+- `python-telegram-bot>=20.0` - Telegram notifications (optional)
+- `GitPython>=3.1.0` - Git integration
+- `docker>=6.0.0` - Docker environment support
+
 **Syntax Highlighting** (enhanced diff display):
 
 ```bash
@@ -263,7 +418,7 @@ Adds:
 **All Optional Features**:
 
 ```bash
-pip install 'claude-dev-cli[generation,toon,plugins]'
+pip install 'claude-dev-cli[generation,toon,plugins,all-pm]'
 ```
 
 **Development Installation** (for contributors):
@@ -770,7 +925,208 @@ cdc history delete 20240109_143022
 - Reduces API costs by ~30-50% in long conversations
 - Shows token savings after summarization
 
-### 7. Usage Tracking
+### 7. Workflows - AI-Powered Automation (v0.7.0+, Enhanced v0.16.1)
+
+Chain multiple AI operations into automated sequences with YAML workflows.
+
+#### Basic Workflow Usage
+
+```bash
+# Run a workflow
+cdc workflow run my-workflow.yaml
+
+# Pass variables
+cdc workflow run workflow.yaml --var file=main.py --var api=local
+
+# List available workflows
+cdc workflow list
+
+# Show workflow details
+cdc workflow show workflow.yaml
+```
+
+#### Example: Simple Review & Test Workflow
+
+```yaml
+name: "Review and Test"
+description: "Review code and generate tests"
+
+steps:
+  - name: code-review
+    command: review
+    args:
+      file: "{{target_file}}"
+      api: "local"  # Use FREE local AI
+      model: "smart-local"
+    output_var: review_result
+  
+  - name: generate-tests
+    command: generate tests
+    args:
+      file: "{{target_file}}"
+      api: "local"
+      model: "code-local"
+    output_var: tests
+  
+  - name: run-tests
+    shell: "pytest tests/"
+    continue_on_error: true
+```
+
+**Run it:**
+```bash
+cdc workflow run review-test.yaml --var target_file=app.py
+```
+
+#### Example: AI Decision Making (NEW in v0.16.1)
+
+Use the `ask` command to make intelligent workflow decisions:
+
+```yaml
+name: "Adaptive Code Analysis"
+description: "AI analyzes code complexity and adapts workflow"
+
+steps:
+  # AI rates complexity
+  - name: check-complexity
+    command: ask
+    args:
+      prompt: "Rate this code complexity 1-10, respond with number only: {{code}}"
+      api: "local"  # FREE
+      model: "fast-local"
+    output_var: complexity
+  
+  # Conditional deep review for complex code
+  - name: deep-review
+    command: review
+    args:
+      file: "{{file}}"
+      api: "personal"  # Use paid API for complex code
+      model: "powerful"
+    if: "{{complexity}} > 7"
+    approval_required: true  # Ask before using paid API
+  
+  # Simple refactor for medium complexity
+  - name: simple-refactor
+    command: refactor
+    args:
+      file: "{{file}}"
+      api: "local"  # FREE
+    if: "{{complexity}} >= 5 and {{complexity}} <= 7"
+```
+
+**Key Features:**
+- ✨ `ask` command for AI queries
+- 💰 Mix free local and paid cloud AI
+- 🎯 Conditional execution based on AI responses  
+- ✋ Approval gates before expensive operations
+
+#### Example: Cost-Optimized Workflow
+
+```yaml
+name: "Smart Cost Optimization"
+description: "Use local AI first, cloud only when critical"
+
+steps:
+  # Quick scan with FREE local AI
+  - name: quick-scan
+    command: ask
+    args:
+      prompt: "Quick scan for obvious issues: {{code}}"
+      api: "local"  # FREE
+      model: "fast-local"
+    output_var: issues
+  
+  # Only use expensive cloud AI if critical issues found
+  - name: deep-security-analysis
+    command: ask
+    args:
+      prompt: "Deep security analysis: {{code}}"
+      api: "personal"  # PAID
+      model: "powerful"  # Claude Opus - expensive
+      system: "You are a security expert"
+    if: "{{issues}} contains 'critical' or {{issues}} contains 'security'"
+    approval_required: true
+  
+  # Tests with local code-specialized model (FREE)
+  - name: generate-tests
+    command: generate tests
+    args:
+      file: "{{file}}"
+      api: "local"  # FREE
+      model: "code-local"
+```
+
+**Cost Savings:**
+- Local AI for initial scans: $0
+- Cloud AI only when needed: saves 70-90%
+- Approval gates prevent accidental costs
+
+#### Workflow Step Types
+
+**1. Command Steps (cdc commands)**
+```yaml
+- name: review-code
+  command: review  # or: generate tests, debug, refactor, etc.
+  args:
+    file: "{{file}}"
+    api: "local"
+    model: "smart-local"
+  output_var: result
+```
+
+**Supported commands:** `review`, `generate tests`, `generate docs`, `refactor`, `debug`, `git commit`, `ask` (v0.16.1)
+
+**2. Shell Steps (any command)**
+```yaml
+- name: run-tests
+  shell: "pytest tests/ -v"
+  output_var: test_output
+  continue_on_error: true
+```
+
+**3. Set Steps (variables)**
+```yaml
+- name: set-config
+  set: api_endpoint
+  value: "https://api.example.com"
+```
+
+#### Advanced Features
+
+**Conditional Execution:**
+```yaml
+if: "{{score}} > 7"  # Simple comparison
+if: "{{result}} contains 'error'"  # String matching
+if: "{{tests.success}}"  # Access step results
+```
+
+**Approval Gates:**
+```yaml
+approval_required: true  # Pause for user confirmation
+```
+
+**Error Handling:**
+```yaml
+continue_on_error: true  # Don't stop workflow on failure
+```
+
+**Variable Interpolation:**
+```yaml
+{{variable}}  # Simple variable
+{{step_name.output}}  # Step output
+{{step_name.success}}  # Step success status
+```
+
+#### Example Workflows Included
+
+**Location:** `examples/workflows/`
+
+1. **review-and-refactor.yaml** - Basic review workflow
+2. **ai-decision-workflow.yaml** - AI-powered decisions (v0.16.1)
+3. **multi-model-workflow.yaml** - Cost optimization (v0.16.1)
+
+### 8. Usage Tracking
 
 ```bash
 # View all usage
@@ -783,7 +1139,7 @@ cdc usage --days 7
 cdc usage --api client
 ```
 
-### 7. TOON Format (Optional - Reduces Tokens by 30-60%)
+### 9. TOON Format (Optional - Reduces Tokens by 30-60%)
 
 ```bash
 # Check if TOON is installed
@@ -952,6 +1308,31 @@ When using a client's Enterprise API:
 |---------|-------------|
 | `cdc git commit` | Generate commit message |
 
+### Workflows (v0.7.0+)
+
+| Command | Description |
+|---------|-------------|
+| `cdc workflow run <file>` | Execute workflow from YAML |
+| `cdc workflow list` | List available workflows |
+| `cdc workflow show <file>` | Show workflow details |
+
+### Ollama (Local AI) (v0.16.0+)
+
+| Command | Description |
+|---------|-------------|
+| `cdc ollama list` | List local models |
+| `cdc ollama pull <model>` | Download model |
+| `cdc ollama show <model>` | Show model details |
+
+### Model Profiles (v0.10.0+)
+
+| Command | Description |
+|---------|-------------|
+| `cdc model list` | List all model profiles |
+| `cdc model show <name>` | Show profile details |
+| `cdc model add <name> <model_id>` | Add custom profile |
+| `cdc model set-default <name>` | Set default profile |
+
 ## Options
 
 ### Common Options
@@ -962,11 +1343,33 @@ When using a client's Enterprise API:
 - `-f, --file <path>`: Include file in prompt
 - `-o, --output <path>`: Save output to file
 
-### Models
+### Model Profiles
+
+**Anthropic (Claude):**
+- `fast` / `claude-3-5-haiku-20241022` - Fast & economical ($0.80/$4 per Mtok)
+- `smart` / `claude-sonnet-4-5-20250929` - Balanced (default) ($3/$15 per Mtok)
+- `powerful` / `claude-opus-4-20250514` - Most capable ($15/$75 per Mtok)
+
+**OpenAI (GPT):** (requires `pip install 'claude-dev-cli[openai]'`)
+- `fast-openai` / `gpt-3.5-turbo` - Fast & cheap ($0.50/$1.50 per Mtok)
+- `smart-openai` / `gpt-4-turbo` - Balanced ($10/$30 per Mtok)
+- `powerful-openai` / `gpt-4` - High capability ($30/$60 per Mtok)
 
-- `claude-3-5-sonnet-20241022` (default)
-- `claude-3-opus-20240229`
-- `claude-3-haiku-20240307`
+**Ollama (Local - FREE):** (requires `pip install 'claude-dev-cli[ollama]'`)
+- `fast-local` / `mistral` - Fast inference (8k context, FREE)
+- `smart-local` / `mixtral` - Powerful reasoning (32k context, FREE)
+- `code-local` / `codellama` - Code specialist (16k context, FREE)
+
+**Usage:**
+```bash
+# Use model profile
+cdc ask -m fast "quick question"  # Uses Haiku
+cdc ask -m fast-local "question"  # Uses Mistral (FREE)
+
+# Use specific model ID
+cdc ask -m claude-opus-4-20250514 "complex task"
+cdc ask -m gpt-4-turbo "openai task"
+```
 
 ## Examples
 
@@ -1015,6 +1418,193 @@ cdc usage --api personal
 cdc usage --api client
 ```
 
+### Project Management & Automation (v0.17.0+)
+
+#### Automated Ticket Execution
+
+```bash
+# Create a ticket
+cdc tickets create "Implement user authentication API" \
+  --description "JWT-based auth with email/password" \
+  --priority high \
+  --type feature
+
+# Output: Created ticket: TASK-A1B2C3D4
+
+# Execute the ticket with full AI automation
+cdc ticket execute TASK-A1B2C3D4 --commit --notify
+
+# What happens:
+# 1. 📥 Fetches ticket and analyzes requirements
+# 2. 🧠 AI generates implementation plan
+# 3. 💻 Generates api/auth.py with login endpoint
+# 4. ✅ Generates tests/test_auth.py with all acceptance criteria
+# 5. 📝 Commits: "feat(TASK-A1B2C3D4): Implement user authentication API"
+# 6. 🎫 Updates ticket status to "completed"
+# 7. 📊 Logs to .cdc-logs/progress.md
+# 8. 🔔 Sends push notification
+
+# Review progress
+cdc log report
+```
+
+#### Bug Reporting with AI Triage
+
+```bash
+# Report a bug - AI automatically classifies it
+cdc bug report \
+  --title "App crashes when email contains + character" \
+  --description "Users with + in email address cannot login" \
+  --expected "All valid email addresses should work" \
+  --actual "Server returns 500 error for emails with +" \
+  --steps "Enter email test+user@example.com" \
+  --steps "Click login" \
+  --steps "App crashes" \
+  --environment production
+
+# AI automatically:
+# ✅ Severity: HIGH (functionality broken)
+# ✅ Category: FUNCTIONALITY
+# ✅ Priority: high
+# ✅ Creates ticket: BUG-XYZ123
+# ✅ Checks for duplicates
+# ✅ Sends notification if critical
+
+# List all bugs
+cdc tickets list --status open
+
+# Execute bug fix
+cdc ticket execute BUG-XYZ123 --commit
+```
+
+#### Progress Tracking
+
+```bash
+# Initialize logging for a project
+cdc log init "E-Commerce Platform"
+
+# Add manual log entries
+cdc log entry "Started authentication module" --level info
+cdc log entry "Completed user registration" --ticket TASK-123 --level success
+cdc log entry "Found issue with password validation" --level warning
+
+# Generate full progress report
+cdc log report
+
+# Output:
+# # Progress Summary
+# 
+# **Total Entries:** 15
+# **Successes:** ✅ 8
+# **Errors:** ❌ 1
+# 
+# ## ✅ 2026-02-01 02:30:15
+# **Ticket:** TASK-123
+# Completed user registration
+# ...
+```
+
+#### Using with repo-tickets Backend
+
+```bash
+# Install repo-tickets integration
+pip install 'claude-dev-cli[all-pm]'
+
+# Navigate to your project with repo-tickets
+cd my-project
+tickets init  # Initialize repo-tickets if needed
+
+# Create a ticket in repo-tickets
+tickets create "Add payment gateway integration"
+# Output: Created TICKET-456
+
+# Execute it with AI
+cdc ticket execute TICKET-456 --backend repo-tickets --commit --notify
+
+# AI will:
+# - Fetch ticket from repo-tickets
+# - Generate payment gateway code
+# - Generate tests
+# - Update repo-tickets status
+# - Commit to your repo
+```
+
+#### Cost Optimization with Local AI
+
+```bash
+# Use FREE local AI (Ollama) for code generation
+# Install Ollama first: https://ollama.ai
+
+# Pull a code model
+cdc ollama pull codellama
+
+# Execute ticket with local AI (zero cost!)
+cdc ticket execute TASK-789 --api local --commit
+
+# Result: Code generated with $0.00 cost!
+```
+
+#### Notifications Setup
+
+```bash
+# Test notifications (uses ntfy.sh - FREE, no signup!)
+cdc notify test --topic my-project
+
+# Check your phone or browser at: https://ntfy.sh/my-project
+# You'll see: "✅ claude-dev-cli notifications are working!"
+
+# All ticket executions can send notifications:
+cdc ticket execute TASK-999 --commit --notify
+
+# Critical bugs auto-notify:
+cdc bug report --title "Database connection fails" \
+  --description "Cannot connect to production DB" \
+  --expected "Should connect" \
+  --actual "Connection timeout" \
+  --environment production
+# 🚨 Sends URGENT notification automatically
+```
+
+#### Complete Workflow Example
+
+```bash
+# 1. Setup
+pip install 'claude-dev-cli[all-pm]'
+cdc log init "My Awesome Project"
+
+# 2. Create tickets for your sprint
+cdc tickets create "User registration" --priority high
+cdc tickets create "User login" --priority high
+cdc tickets create "Password reset" --priority medium
+cdc tickets create "Email verification" --priority low
+
+# 3. Execute tickets one by one
+cdc ticket execute TASK-001 --commit --notify
+cdc ticket execute TASK-002 --commit --notify
+
+# 4. Bug is reported
+cdc bug report \
+  --title "Registration fails for Gmail users" \
+  --description "Gmail addresses rejected" \
+  --expected "Should accept all emails" \
+  --actual "Validation error"
+
+# 5. Fix the bug
+cdc ticket execute BUG-001 --commit --notify
+
+# 6. Review progress
+cdc log report
+cdc tickets list
+
+# 7. All done! 🎉
+# - Code generated
+# - Tests written
+# - Tickets updated
+# - Commits made
+# - Team notified
+# - Progress logged
+```
+
 ## Environment Variables
 
 | Variable | Description |