<div align="center">
<picture> <source media="(prefers-color-scheme: dark)" srcset="docs/images/brand/anolisa-lockup-dark.svg" > <source media="(prefers-color-scheme: light)" srcset="docs/images/brand/anolisa-lockup-light.svg" > <img src="docs/images/brand/anolisa-lockup-light.svg" alt="ANOLISA" width="320" > </picture>
<sub>Agentic Nexus Operating Layer & Interface System Architecture</sub>
The operating system layer for Agent workloads.
Let Agents drive the system straight from your terminal, and strip the tool responses that reach the model before they cost you — while keeping the Shell, Agent framework, and sandbox you already run.
中文版 · Website · Quick Start · User Guide · Contributing
 
</div>
---
ANOLISA is a server-side operating layer for AI Agent workloads. It addresses three practical constraints of Agent execution: terminal entry, Token cost, and execution environments. Keep the Shell, Agent framework, and sandbox you already use. ANOLISA CLI provides a single installation entry point, while each capability can be enabled independently.
<p align="center"> <img src="docs/images/readme/highlights.png" alt="ANOLISA product highlights" /> </p>
What it solves
<p align="center"><strong>01 · AGENT INTERFACE</strong></p>
<h3 align="center">Let the Agent work directly in the terminal</h3>
cosh-ng gives Agents a structured, predictable interface for Shell and system operations. Keep the Agent framework and sandbox already in use, and bring system work into the existing terminal workflow.
<p align="center"><strong>02 · CONTEXT EFFICIENCY</strong></p>
<h3 align="center">See where Tokens go and cut waste before it reaches the model</h3>
Token-less removes redundancy from tool schemas and responses. Agent Memory reuses context across sessions, SkillFS exposes Skills as views and mounts them on demand so only the relevant ones enter the context, and AgentSight records where Tokens are actually spent.
<p align="center"> <img src="docs/images/readme/tokenless-response.png" alt="Token-less response compression in the terminal" /> </p>
debug and trace are dropped by the field blacklist, metadata as null, and
tags / extra as empty values. Compression runs between the Agent and the model, so no Agent framework code changes. Dropped array items stay retrievable through a <<tokenless:KEY>> marker, which keeps the compression reversible.
| Tool responses | Tool schemas | Full pipeline | |----------------|--------------|---------------| | 65.8% fewer Tokens | 47.3% fewer Tokens | 62.9% fewer Tokens | | ResponseCompressor · 46.85 µs | SchemaCompressor · 11.44 µs | 198.91 µs |
Savings apply to the tool responses entering the context, not to the whole session bill — the Token-less README explains how to estimate the effect for a given workload.
Read the Token-less user manual →
<p align="center"><strong>03 · EXECUTION RUNTIME</strong></p>
<h3 align="center">Give every Agent execution a boundary and a way back</h3>
ANOLISA is building out the Agent execution environment: Agent Sec Core isolates risky operations, and ws-ckpt keeps recovery points for workspace changes.
Install
ANOLISA CLI is the common installation entry point. Enable cosh-ng, Token-less, or other capabilities as needed.
curl -fsSL https://get.agentic-os.sh | bash
anolisa install cosh-ng
anolisa install tokenlessRun cosh-ng to enter the Agent-native shell, or point any existing Agent at it — Token-less applies to tool calls without further configuration.
Documentation
Quick Start · Installation · User Guide · Troubleshooting · Build from Source · Changelog
Community
<div align="center">
<img src="docs/images/readme/dingtalk-qr.png" alt="ANOLISA DingTalk community QR code" width="180"/>
Scan with DingTalk to join the ANOLISA community.
</div>
- Open an issue for bugs and
feature requests.
- Read CONTRIBUTING.md before submitting a pull request.
- Report vulnerabilities through the Security Policy.
License
ANOLISA is released under the Apache License 2.0.
