coru.dev · official home of the Koru OSS library
Koru
Open-source closed-loop refactor orchestration
pip install koru
koru autoHow It Works
Closed-loop refactor on your repo
Python package on PyPI that coordinates LLM-augmented refactor tools with ticket-driven workflow
and regression-free verification across multi-repo workspaces.
Detect → Plan → Execute → Verify → Heal → Learn → Repeat
Core Mission
Transform how teams approach large-scale code modernization with intelligent, auditable automation.
Closed-Loop Automation
Detect issues → plan fixes → execute changes → verify safety → heal failures → repeat
Ticket-Driven Workflow
Structured, auditable refactor pipelines with zero-trust policy gates
Regression-Free Verification
Multi-tier validation (unit tests, integration tests, semantic analysis)
LLM Orchestration
Coordinate multiple AI agents (Windsurf, Cursor, Claude Code, aider, OpenRouter)
What Koru Delivers
Capabilities shipped in the open-source library — discovery, ticketed execution, IDE automation, verification, and fleet-scale rollout from one CLI.
Discovery Passes
Broad codebase analysis that turns signals into concrete planfile tickets
Ticketed Delivery
Planfile-first execution so every change is scoped, auditable, and resumable
IDE / Daemon Automation
Cursor, VS Code, Windsurf, and plugin-driven hands-off refactor loops
Verification & Healing
Post-run checks, runtime probes, and recovery paths that keep regressions visible
Multi-Repo Rollout
One workflow across matching repositories, from local repair to fleet-wide rollout
Auditing & Traceability
JSONL event logs, planfile handoffs, and machine-readable history for every run
The Koru Loop
Connected Ecosystem
20+ integrated tools for the complete refactor automation pipeline
planfile
Ticket backlog & execution gateway
regix
Python regression metrics & tracking
redup
Duplicate detection & deduplication
testql
Declarative HTTP scenario tests
wup
Real-time service dependency health
vallm
Multi-tier semantic patch validator
redsl
OpenRouter-powered auto-refactor
prefact
Proactive LLM-aware linter
costs
Zero-config AI cost per commit
Koru in Action
Real situations, real terminal output. Pick a scenario — the loop drives an external IDE, a vendor CLI, or a headless LLM, and always verifies before it commits.
Quick Start
Two commands to install and run Koru on your repo. For wizard, queue mode, scan passes, and other workflows, see the README and CLI examples.
pip install koru
koru autoOperational Modes
Executor: Windsurf/Cursor LLM + regix/pytest
Executor: OpenRouter (redsl, vallm, llx)
Executor: Both modes + autopilot injection
By The Numbers
Try Koru on your repo
Install from PyPI and run the closed loop on your workspace — see Quick Start for the exact commands.
Open Source Library
Koru (github.com/semcod/koru) is the maintained OSS project behind this site — not a separate commercial product. Install from PyPI, read the docs on GitHub, and contribute under Apache-2.0.
Package
Base CLI with planfile queue, scan, and autopilot protocol. Optional extras cover API, agent, and dev workflows.
First run
Install from PyPI, then run the closed loop on your repo — workspace setup, scan, and ticket execution in one flow. See Quick Start for the two-command sequence.
Source & docs
README · docs/ · PyPI releases
Validated on semcod monorepos, sample demo workspaces, and other multi-repo setups.