now:
before:
- π€ LLM usage: $7.5000 (96 commits)
- π€ Human dev: ~$3835 (38.3h @ $100/h, 30min dedup)
Generated on 2026-05-03 using openrouter/qwen/qwen3-coder-next
Planfile is an SDLC automation platform that provides strategic project management with CI/CD integration and automated bug-fix loops. It manages sprints and strategies across external ticket systems like GitHub, Jira, and GitLab.
- 56 modules with 395 functions
- Cyclomatic Complexity: CCΜ=4.1 (improved from 4.2)
- Critical functions: 15 (target: β€4)
- Zero circular dependencies
- Languages: Python (53), Shell (17), JavaScript (3)
- π― Strategy Modeling: Define strategies and sprints in YAML with task patterns
- π Automated CI/CD Loop: Test β Ticket β Fix β Retest automation
- π Multi-Backend Support: Integrates with GitHub Issues, Jira, GitLab, and generic HTTP APIs
- π€ LLM-Powered: AI-driven bug reports and auto-fix capabilities
- π Progress Tracking: Review strategy execution with detailed metrics
- π CLI Tool: Easy-to-use command-line interface for applying and reviewing strategies
- π Python Library: Use planfile programmatically in your Python applications
- π REST API: Run as FastAPI server for HTTP access and integrations
- π¨ Rich Output: Beautiful terminal output with progress bars and tables
- π³ Docker Support: Containerized deployment with Ollama integration
- π§ Extensible: Easy to add new backends and custom integrations
- π Code Analysis: Integration with external tools (code2llm, vallm, redup)
- π MCP Server: Model Context Protocol server integration
- π€ LLX Integration: Advanced code analysis and model selection
- π Proxy Routing: Smart model routing via Proxym API
- π Metrics-Driven: Project metrics analysis for informed planning
- π£οΈ DSL (Domain Specific Language): Natural language-like commands for planfile operations
pip install planfile
pip install planfile[all]
pip install planfile[github,jira] pip install planfile[gitlab]
### 1. Create a Strategy
Create a `strategy.yaml` file:
```yaml
name: "My Project Strategy"
project_type: "web"
domain: "fintech"
goal: "Launch a secure payment processing platform"
sprints:
- id: 1
name: "Core Infrastructure"
length_days: 14
objectives:
- Set up project structure
- Implement authentication
tasks:
- type: "feature"
title: "Setup project structure"
description: "Create basic project layout and configuration"
estimate: 2
priority: "high"
- id: 2
name: "Payment Processing"
length_days: 21
objectives:
- Implement payment gateway
- Add security measures
tasks:
- type: "feature"
title: "Payment gateway integration"
description: "Integrate with payment provider API"
estimate: 5
priority: "critical"
export GITHUB_TOKEN=your_token export GITHUB_REPO=owner/repo
export JIRA_URL=https://company.atlassian.net export JIRA_EMAIL=your@email.com export JIRA_TOKEN=your_token export JIRA_PROJECT=PROJ
export GITLAB_TOKEN=your_token export GITLAB_PROJECT_ID=123
# Run automated bug-fix loop
planfile auto loop \
--strategy ./strategy.yaml \
--project . \
--backend github \
--max-iterations 5 \
--auto-fix
# Check CI status
planfile auto ci-status
# Review strategy progress
planfile strategy review \
--strategy ./strategy.yaml \
--project . \
--backend github
from planfile import Planfile, quick_ticket
# Auto-discover .planfile/ in your project
pf = Planfile.auto_discover(".")
# Create tickets programmatically
ticket = pf.create_ticket(
title="Fix authentication bug",
description="Users cannot login with OAuth",
priority="high",
labels=["bug", "backend"]
)
# List and filter tickets
tickets = pf.list_tickets(sprint="current", status="open")
# Quick one-liner for tools
ticket = quick_ticket("Production alert", tool="monitoring", priority="critical")uvicorn planfile.api.server:app --reload --port 8000
curl "http://localhost:8000/tickets?sprint=current"
curl -X POST "http://localhost:8000/tickets"
-H "Content-Type: application/json"
-d '{"title": "API fix", "priority": "high"}'
### 5. Ticket Management CLI
Create, update, delete and bulk-manage tickets directly from command line:
```bash
# Create a new ticket
planfile ticket create "Fix login bug" -p high -l bug -l backend
# List tickets with filters
planfile ticket list --status open
planfile ticket list --label bug --format json
# Update a single ticket
planfile ticket update PLF-001 --status done
planfile ticket update PLF-002 --priority critical
# Validate if tickets are still current
planfile ticket validate
planfile ticket validate Q01 Q02 --issues analyses/issues.yaml --format yaml
planfile ticket validate --stale-only --fail-on-stale
# Delete tickets by ID
planfile ticket delete PLF-001 PLF-002
# Delete tickets matching filters (with confirmation)
planfile ticket delete --status done --force
planfile ticket delete --sprint old-sprint --label archive
# Preview deletions without executing
planfile ticket delete --label old --dry-run
# Bulk update tickets by filters
planfile ticket bulk-update --status-filter open --new-status in_progress
planfile ticket bulk-update --label todo --new-status done --force
# Change priority for all open bugs
planfile ticket bulk-update --status-filter open --label bug --new-priority high
# Add/remove labels in bulk
planfile ticket bulk-update --label old --add-label archived --remove-label old
# Move tickets between sprints
planfile ticket bulk-update --sprint current --status-filter done --move-to-sprint completed
# Combine multiple updates
planfile ticket bulk-update \
--sprint sprint-1 \
--status-filter open \
--new-status in_progress \
--new-priority high \
--add-label urgent
Auto-sync after changes:
# Create ticket and sync immediately
planfile ticket create "Fix login bug" -p high --sync
# Delete tickets and auto-sync
planfile ticket delete --status done --sync
# Bulk update with sync
planfile ticket bulk-update --label bug --new-status done --sync
# Preview sync (dry-run)
planfile ticket update PLF-001 --status done --sync --sync-dry-runSynchronize tickets with TODO.md, GitHub, Jira, and GitLab:
# Sync with markdown files (TODO.md, CHANGELOG.md)
planfile sync markdown
# Preview sync without making changes
planfile sync markdown --dry-run
# Sync only to external system (export)
planfile sync markdown --direction to
# Sync only from external system (import)
planfile sync markdown --direction from
# Sync with GitHub Issues (requires github.planfile.yaml config)
planfile sync github
# Sync with all configured integrations
planfile sync all
# Example: Export all done tickets to CHANGELOG.md
planfile ticket bulk-update --status-filter done --add-label changelog
planfile sync markdown --direction to --dry-run
# Watch .planfile/ directory and auto-sync on changes
planfile sync watch
# Watch with custom polling interval (seconds)
planfile sync watch --interval 10
# Watch specific integrations only
planfile sync watch --integration github --integration jira
# One-time sync (no watch loop)
planfile sync watch --once
# Watch with sync direction "from" (import from external systems)
planfile sync watch --direction fromSync Configuration:
Create github.planfile.yaml for GitHub integration:
integrations:
github:
repo: "owner/repo"
token: "${GITHUB_TOKEN}" # or auto-detected from `gh auth token`Markdown sync works out-of-the-box with TODO.md and CHANGELOG.md files.
To enable automatic checkbox syncing from planfile.yaml task statuses/results
in tools that reuse planfile APIs (e.g. llx plan run), add:
integrations:
markdown:
sync_on_plan_run: true
todo_file: TODO.mdPlanfile includes a natural language-like DSL for quick operations:
# CLI DSL - single command
planfile dsl "list tickets sprint=current status=open"
planfile dsl 'create ticket "Fix login bug" priority=high'
planfile dsl "update ticket PLF-001 status=done"
planfile dsl "move ticket PLF-001 to sprint=2"
planfile dsl "done ticket PLF-001"
planfile dsl "validate"
planfile dsl "sync github"
# CLI DSL - interactive shell
planfile dsl
# Then type commands like:
# > list tickets sprint=current
# > create ticket "New task" priority=high
# > exit
# Format options
planfile dsl "list tickets" --format json
planfile dsl "list tickets" --format yamlPython API DSL:
from planfile import DSLExecutor
executor = DSLExecutor(project_path=".")
result = executor.run("list tickets sprint=current")
print(result.ok, result.data)
# Parser only
from planfile import DSLParser
parser = DSLParser()
cmd = parser.parse('create ticket "Fix bug" priority=high')
print(cmd.verb, cmd.target, cmd.params)REST API DSL:
# Start server
uvicorn planfile.api.server:app --reload
# Execute DSL via HTTP
curl -X POST "http://localhost:8000/dsl" \
-H "Content-Type: application/json" \
-d '{"command": "list tickets", "project_path": "."}'
# WebSocket DSL
ws://localhost:8000/ws?project_path=.MCP DSL Tool:
# Start MCP server
python -m planfile.mcp.server
# MCP tool call (JSON-RPC)
{
"jsonrpc": "2.0",
"id": 1,
"method": "tools/call",
"params": {
"name": "planfile_dsl",
"arguments": {
"command": "list tickets sprint=current",
"project_path": "."
}
}
}Supported DSL Commands:
list ticketsβ list with filters (sprint, status, priority, label)create ticketβ create with name, priority, sprint, labelsshow ticketβ show ticket by IDupdate ticketβ update status, priority, name, labelsmove ticketβ move to another sprintdone ticketβ mark as donestart ticketβ mark as in_progressblock ticketβ mark as blockeddelete ticketβ delete ticketlist sprintsβ list all sprintsadd sprintβ add new sprintvalidateβ validate ticketssyncβ sync to integrations (github, jira, gitlab, all)queryβ query tickets with where clauseexportβ export to formats (json, yaml, html)helpβ show command reference
Aliases: add/new β create, ls β list, get/show β show, set/edit/patch β update, mv β move, finish/complete β done, begin β start
make ci-loop STRATEGY=strategy.yaml BACKENDS=github MAX_ITERATIONS=5
make example-github make example-jira
make pipeline-docker
## π CI/CD Automation
Planfile provides complete automation for the bug-fix lifecycle:
1. **Test Execution**: Run your test suite
2. **Bug Detection**: Identify failing tests and code issues
3. **AI Analysis**: Generate detailed bug reports using LLM
4. **Ticket Creation**: Create tickets in your PM system
5. **Auto-Fix**: Optionally fix bugs automatically with AI
6. **Verification**: Re-run tests to verify fixes
7. **Loop**: Repeat until all tests pass
# Run with Docker Compose
docker-compose up -d
# Run auto-loop in container
docker-compose exec sprintstrat-runner planfile auto loop \
--strategy /app/strategy.yaml \
--project /workspace \
--backend github \
--max-iterations 5
- Documentation Navigation
- CI/CD Integration Guide
- API Reference
- CLI Reference
- Examples
- Architecture Overview
- Migration Guide
- Performance Guide
- Examples Runner Guide
- Standalone Usage Guide
- Changelog
The strategy.yaml file supports:
- Sprints: Time-boxed development periods with embedded tasks
- Goals: Project objectives with success criteria
- Quality Gates: Definition of done criteria
- Model Hints: AI model suggestions for different phases
name: "My Project Strategy"
project_type: "web"
domain: "fintech"
goal:
title: "Launch secure payment platform"
description: "Build and deploy a secure payment processing system"
success_metrics:
- "99.9% uptime"
- "<100ms response time"
sprints:
- id: 1
name: "Core Infrastructure"
length_days: 14
objectives: ["Set up project structure", "Implement authentication"]
tasks:
- type: "feature"
title: "Setup project structure"
description: "Create basic project layout and configuration"
estimate: 2
priority: "high"
model_hints:
small: "gpt-5.4-mini"
large: "gpt-4o"
quality_gates:
- name: "Code Coverage"
criteria: ["coverage >= 80%"]
required: trueEach backend requires specific configuration:
# GitHub backend
github:
token: ${GITHUB_TOKEN}
repo: ${GITHUB_REPO}
# Jira backend
jira:
url: ${JIRA_URL}
email: ${JIRA_EMAIL}
token: ${JIRA_TOKEN}
project: ${JIRA_PROJECT}Planfile integrates with multiple LLM services:
- Multiple Providers: OpenAI, Anthropic, LiteLLM, Local LLMs
- Smart Routing: Automatic model selection via Proxym proxy
- Code Analysis: LLX integration for advanced metrics
- Auto-Fix: Automatic code generation for bug fixes
- Strategy Generation: AI-powered strategy creation
# Enable AI features
export OPENAI_API_KEY=your_key
export ANTHROPIC_API_KEY=your_key
export PROXY_API_URL=http://localhost:9999
# Run with auto-fix
planfile auto loop --strategy strategy.yaml --auto-fix
# Generate strategy with AI
planfile strategy generate ./my-project --model gpt-4oExplore the examples/ directory for comprehensive use cases:
planfile examples list
planfile examples run code2llm
planfile examples run --all
### Featured Examples
- **[checkbox-tickets](examples/checkbox-tickets/)** - Native markdown checkbox support (`- [ ]` / `- [x]`)
- **[code2llm](examples/code2llm/)** - Code analysis with LLM integration
- **[bash-generation](examples/bash-generation/)** - Generate bash scripts from strategies
- **[cli-commands](examples/cli-commands/)** - CLI usage patterns
- **[advanced-usage](examples/advanced-usage/)** - CI/CD integration examples
- **[interactive-tests](examples/interactive-tests/)** - Interactive mode demonstrations
- **[ecosystem](examples/ecosystem/)** - MCP, LLX, and proxy routing integrations
- **[cli](examples/cli/)** - DSL command-line interface examples
- **[mcp](examples/mcp/)** - MCP DSL tool integration
- **[python-api](examples/python-api/)** - Python API DSL usage
- **[rest-api](examples/rest-api/)** - REST API and WebSocket DSL
# examples/quick-start.yaml
name: "Quick Start Demo"
project_type: "web"
domain: "demo"
goal: "Demonstrate planfile capabilities"
sprints:
- id: 1
name: "Setup"
length_days: 7
tasks:
- type: "feature"
title: "Initialize project"
description: "Create basic project structure"
estimate: 1
priority: "high"
For more examples, see the examples directory.
name: "E-commerce MVP"
project_type: "web"
domain: "ecommerce"
goal: "Launch minimum viable e-commerce platform"
sprints:
- id: 1
name: "Foundation"
length_days: 10
tasks:
- type: "feature"
title: "Setup project structure"
estimate: 1
- type: "feature"
title: "Database schema"
estimate: 3name: "Mobile App MVP"
project_type: "mobile"
domain: "healthcare"
goal: "Launch health tracking mobile app"
sprints:
- id: 1
name: "Core Features"
length_days: 14
tasks:
- type: "feature"
title: "User authentication"
estimate: 3
- type: "feature"
title: "Health data tracking"
estimate: 5- GitHub: Issues, Projects, Actions
- GitLab: Issues, CI/CD, Merge Requests
- Bitbucket: Issues, Pipelines
- Jira: Issues, Epics, Sprints
- Linear: Issues, Projects, Teams
- ClickUp: Tasks, Lists, Spaces
- Asana: Tasks, Projects, Portfolios
- GitHub Actions: Workflow automation
- GitLab CI: Pipeline integration
- Jenkins: Build automation
- Azure DevOps: Release management
git clone https://github.com/semcod/planfile cd strategy
python -m venv .venv source .venv/bin/activate # On Windows: .venv\Scripts\activate
pip install -e ".[dev]"
ruff check src/ tests/ ruff format src/ tests/
### Project Structure
planfile/ βββ planfile/ # Main package β βββ analysis/ # Code analysis components β β βββ external_tools.py # External tool integrations β β βββ generator.py # Strategy generation β β βββ file_analyzer.py # File analysis β β βββ sprint_generator.py # Sprint generation β β βββ parsers/ # YAML/JSON/Toon parsers β β βββ generators/ # Metrics extractors β βββ cli/ # CLI commands β β βββ cmd/ # Core commands (init, generate, apply, etc.) β β βββ auto_loop.py # CI/CD automation β β βββ extra_commands.py # Additional utilities β βββ integrations/ # Backend integrations β β βββ github.py # GitHub Issues β β βββ jira.py # Jira β β βββ gitlab.py # GitLab β β βββ generic.py # Generic HTTP API β βββ llm/ # LLM integrations β β βββ adapters.py # Multiple LLM adapters β β βββ client.py # LLM client β β βββ generator.py # Strategy generation β βββ loaders/ # Data loaders β β βββ yaml_loader.py # YAML loading/saving β β βββ cli_loader.py # JSON/Markdown export β βββ models.py # Core data models β βββ models_v2.py # Simplified models β βββ runner.py # Strategy execution β βββ ci_runner.py # CI/CD automation β βββ executor_standalone.py # Standalone executor βββ examples/ # Usage examples β βββ quick-start/ # Basic examples β βββ ecosystem/ # Integration examples β βββ advanced-usage/ # Advanced features βββ tests/ # Test suite βββ docs/ # Documentation βββ project/ # Project analysis output
## π License
This project is licensed under the Apache License 2.0 - see the [LICENSE](LICENSE) file for details.
## π€ Contributing
Contributions are welcome! Please feel free to submit a Pull Request. For major changes, please open an issue first to discuss what you would like to change.
## π Support
- π [Documentation](docs/)
- π [Issue Tracker](https://github.com/semcod/planfile/issues)
- π¬ [Discussions](https://github.com/semcod/planfile/discussions)
## π Acknowledgments
- Built with [Typer](https://typer.tiangolo.com/) for CLI
- Styled with [Rich](https://rich.readthedocs.io/) for terminal output
- Validated with [Pydantic](https://pydantic-docs.helpmanual.io/) for data models
---
**Planfile** - Your strategic partner in SDLC automation. π
## License
Licensed under Apache-2.0.
## Status
_Last updated by [taskill](https://github.com/oqlos/taskill) at 2026-04-25 13:42 UTC_
| Metric | Value |
|---|---|
| HEAD | `f7a4f3c` |
| Coverage | β |
| Failing tests | β |
| Commits in last cycle | 50 |
> Introduces a configuration management system across goal, examples, and docs; refactors and extends the code analysis engine and test suite to improve coverage. Includes documentation fixes (markdown output), a fix for code quality metrics, and cleanup removing large generated files.
<!-- taskill:status:end -->
<!-- taskill:status:start -->
## Status
_Last updated by [taskill](https://github.com/oqlos/taskill) at 2026-04-25 18:22 UTC_
| Metric | Value |
|---|---|
| HEAD | `849565a` |
| Coverage | β |
| Failing tests | β |
| Commits in last cycle | 9 |
> Implemented a comprehensive configuration management system and CLI interface for the planfile project, including new command groups for backlog, sync, ticket, and validate operations.
<!-- taskill:status:end -->

