
YYLO
yylo.devAgentic engineering coordination with durable evidence
Coordinate coding agents from intent through validated code while preserving task, session, review, and evaluation evidence.

Overview
YYLO is an open-source system of work for agentic engineering, coordinating coding agents from engineering intent through validated code while preserving task, session, review, and evaluation evidence. It provides a control plane for action, a Git-native Ledger for task truth, and Benchmark evidence for comparing agent attempts. YYLO is for software organizations using multiple AI tools, including engineering leaders, platform teams, and technical founders.
How it works
YYLO coordinates agent execution without replacing Git or the repository workflow. It defines intent with a task that records one bounded outcome, constraints, and dependencies. It delegates execution to an agent while preserving its worktree, model, and session provenance. Validation uses tests, evidence receipts, and policy-bound review to make acceptance inspectable. Task responses, commits, review evidence, and merge outcomes remain available for future work.
Key capabilities
YYLO includes three open-source layers: the control plane coordinates action, the Ledger preserves task truth in Git, and Benchmark turns attempts into evaluation evidence. The control plane delegates bounded agent execution and integrates accepted changes. The Ledger keeps intent, status, dependencies, responses, archives, and commit evidence durable. Benchmark reviews cases, delegates independent attempts, and evaluates retained outputs with separate checks and judge assessments.
Who it is for
YYLO is for software organizations using multiple AI tools, including engineering leaders, platform teams, developer-productivity teams, and technical founders. It helps teams coordinate agents, preserve shared knowledge, and evaluate outcomes across users, workflows, and AI providers. The platform is designed to become the trusted context layer for people, agents, and business systems. It aims to make AI-assisted work repeatable, visible, and governable.
Pricing and limits
YYLO is open-source and available today, with three working, publicly available products. The current stable versions are YYLO 0.2.11, Ledger 0.4.0, Benchmark 0.2.1, and Skills 2.1.1. Installation is via npm with 'npm install -g @yylo/cli@0.2.11'. The project is in early stages, with future paid team offerings and pricing under validation. The source states that paid capabilities are direction, not current traction.
Key features
Agent coordination
Coordinate bounded agent execution with provenance, validation, review, and integration.
Git-native Ledger
Preserve task truth, dependencies, responses, archives, and commit evidence in Git.
Benchmark evaluation
Compare agent attempts with retained benchmark evidence and frozen checklists.
Skills for agents
Reusable procedures for Ledger Records, planning, and bounded delivery.
Multi-provider support
Provider-independent coordination across models and agents.
Use cases
Agentic engineering workflows
Coordinate coding agents from intent through validated code with evidence.
Engineering memory
Preserve task truth, dependencies, and commit evidence in Git.
Agent evaluation
Compare agent attempts with retained benchmark evidence.
Pricing & limits
From the official site · checked 10/10/2026Frequently asked questions about YYLO
- What is YYLO?
- YYLO is an open-source system of work for agentic engineering. It coordinates coding agents from intent through validated code while preserving task, session, review, and evaluation evidence.
- Who is YYLO for?
- YYLO is for software organizations using multiple AI tools, including engineering leaders, platform teams, developer-productivity teams, and technical founders.
- Is YYLO free?
- Yes, YYLO is open-source and available today. The source mentions 'Open-source, product-led entry' and 'Three working, publicly available open-source products.'
- How do I install YYLO?
- YYLO is installed via npm with the command 'npm install -g @yylo/cli@0.2.11'. It works with Git and supports multiple AI providers.
- What are the main components of YYLO?
- YYLO includes the YYLO CLI for coordination, YYLO Ledger for Git-native task truth, YYLO Benchmark for evaluation evidence, and YYLO Skills for reusable procedures.
Related tools

Test-Lab.ai
Write tests in plain English and get results in minutes with autonomous AI agents that navigate your app like real users.

Clipy: Agent-Readable Screen Recorder for Coding Agents
Record bugs and UI feedback once; coding agents get context they can act on.

Affordable LLM API Gateway · GLM, Kimi, DeepSeek - A2Agent
Access DeepSeek, GLM, Kimi, Qwen, and MiniMax models at discounted rates through one API key.

Kintsu.ai
Edit your WordPress site by describing changes in plain English, with safe previews and one-click publishing.

Wallarm
Discover, observe, enforce, and govern AI workloads on AWS with kernel-level visibility and no code changes.

Code Guardian
Enforce quality gates and safety latches on AI-generated code in Claude Code and Codex.

AICost
Independent LLM API price desk: compare model pricing across OpenAI, Anthropic, Google and more.

PineGen AI
Generate, debug, and backtest TradingView-ready Pine Script v6 strategies with AI.