OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Agent Infrastructure

Anolisa

ANOLISA is a server-side operating layer for AI agent workloads, delivered through a CLI with independently enabled components for terminal interaction, tool-response compression, runtime and security, and observability.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Anolisa — OpenViST project profile in Agent Infrastructure.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use ANOLISA when you want a self-hosted operating layer around agent workloads, especially to reduce model-bound tool output, observe agent runs or add runtime and security components without replacing your existing shell, framework or sandbox.

Who it is for

  • Developers and teams operating AI agents who want to add workload controls while keeping their existing shell, agent framework and sandbox.
  • Terminal users who want agents to work alongside familiar Bash or Zsh behavior.
  • Teams seeking to reduce tool-response content passed to models or inspect agent activity and token use.

Documented uses

  • Add a CLI entry point for managing agent workload capabilities.
  • Compress tool schemas and responses before they reach a model.
  • Trace agent activity and token use; on Linux, AgentSight uses eBPF without changing agent code.
  • Reuse useful context across sessions and keep the current skill view focused.
  • Add runtime features such as checkpointing and rollback, sandbox lifecycle management, and sandboxing and verification.
Before you choose

Requirements & limits

The project repository is public and lists an Apache-2.0 license. The README lists Linux and macOS as platforms; AgentSight's eBPF tracing is described for Linux.

  • ANOLISA is an infrastructure layer with separately enabled components, not just a terminal agent interface.
  • The documented Token-less example connects to Claude Code and includes a restart before running a task.
  • AgentSight's eBPF-based observation is specifically described for Linux.
  • The repository identifies Rust as its primary language and Apache-2.0 as its license.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set up Token-less with Claude Code

Install Token-less, enable its Claude Code adapter and inspect its statistics after a tool-heavy task.

Before you start

  • ANOLISA CLI installation entry point documented in the project README.
  • Claude Code, for the adapter workflow described by the README.
  • A terminal to run the documented commands.

Steps

  1. Run the documented installer: the command below.

    Documented command
    curl -fsSL https://get.agentic-os.sh | bash

  2. Add $HOME/.local/bin to PATH for the current shell with export PATH="$HOME/.local/bin:$PATH".

  3. Run anolisa install tokenless and anolisa adapter enable tokenless claude-code.

  4. Restart Claude Code and run a tool-heavy task.

  5. Review results with tokenless stats summary and tokenless stats list --limit 5.

Expected outcome

Token-less is installed and enabled for Claude Code, and its documented statistics commands can be used to inspect results after a task.

Guide limitations

  • The documented sequence requires restarting Claude Code before the task.
  • The README presents this as a Token-less and Claude Code workflow; it does not establish that the same steps apply to other agent frameworks.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
13 October 2026
Original website title
agentic-os-org/ANOLISA · Official homepage title · retained in its source language
How we review sourcesSuggest a correction