Get up and running with Loop Engineering in 5 minutes.
git clone https://github.com/chillum-codeX/loop-engineering.git
cd loop-engineering
pip install -e .Or install the v0.4.1 release wheel directly:
pip install https://github.com/chillum-codeX/loop-engineering/releases/download/v0.4.1/loop_engineering-0.4.1-py3-none-any.whlThe package is not yet published to PyPI.
loop-engine init --pattern daily-triage --name my-first-loop
cd my-first-loopThis creates:
my-first-loop/
|-- loop.yaml # Main configuration
|-- README.md # Project docs
|-- .gitignore # Git ignore rules
`-- .loop/
|-- skills/ # SKILL.md files
|-- state/ # Persistent state
`-- worktrees/ # Git worktrees
Edit loop.yaml:
name: my-first-loop
pattern: daily-triage
tool: claude
runtime:
max_iterations: 10
checkpoint_preset: manual
budget:
max_tokens: 10000
max_cost: 1.00
max_steps: 10loop-engine auditYou should see:
Audit Results
Score: 100/100
[####################] 100%
Test without making changes:
loop-engine run --dry-runloop-engine runFor programmatic use:
import asyncio
from loop_engine import RuntimeConfig, create_runtime
config = RuntimeConfig()
config.discovery.skills_dir = ".loop/skills"
config.persistence.state_dir = ".loop/state"
config.persistence.format = "sqlite"
config.handoff.default_token_budget = 100_000
config.handoff.default_cost_budget = 10.0
config.handoff.default_step_budget = 50
runtime = create_runtime(runtime_config=config, max_iterations=50)
result = asyncio.run(runtime.run())
print(f"Completed: {result.status.name}")
print(f"Tasks completed: {result.tasks_completed}")- Try the starter patterns
- Read the state machine specification
- Review the step lifecycle specification
- Understand the recovery model