ai-agent

mirage: Seamless Virtual Filesystem for Your AI Agents

mirage lets founders give AI agents a unified file system, simplifying data access across languages and tools. Backed by 3.7k+ GitHub stars, it’s ideal for teams building LLM‑powered workflows.

From the repo · “The World's First Unified Virtual Filesystem For AI Agents”

3,721 stars+25★ this week274 forksTypeScriptGuide quality 9/10 · reviewed Oct 6Updated 10/24/2026100% free · open source
What it does

Mirage provides a unified virtual filesystem for AI agents, allowing them to interact with multiple data sources in a single, cohesive environment.

When to use it
  • •When building AI agents that need to access and manipulate diverse data sources
  • •When requiring a unified interface for AI agents to read and write data across different systems
  • •When needing to simplify data management for AI agents with multiple, disparate data sources
Ready-to-paste prompt
npm run start && node examples/agent-example.ts
Heads up: Make sure you have Node.js 16 or later installed, as mirage relies on features only available in this version or newer
Saves to your device
Use with Claude
New

Skip the builder — one click puts this in Claude, Cursor, Antigravity and more.

🛠️ Technical setup

Expect 20–40 minutes in a terminal — or let your AI agent drive it.

Try it instantly — no install
Claude Code
mkdir -p ~/.claude/skills/mirage && curl -fsSL https://workflowstacks.com/api/skills/mirage/claude-skill -o ~/.claude/skills/mirage/SKILL.md
Open in another AI app

Opens the app with this repo with the prompt ready to go — no copy-paste needed.

Connect the whole catalog (MCP)
claude mcp add --transport http workflowstacks https://workflowstacks.com/api/mcp

Adds a WorkflowStacks connector to Claude Code: search and load any skill here by chatting.

How mirage: Seamless Virtual Filesystem for Your AI Agents works
Codeflow
Free to inspect

Mirage: Seamless Virtual Filesystem for Your AI Agents is a very large TypeScript project (~891k lines across 6430 code files, plus 1891 test files). It is a full software project: use it through its install path rather than reading it end to end. Last commit this month, Apache-2.0 license, has a test suite.

Size
Very large codebase
~891k lines · 6430 code files · days to read — use, don't read
Setup
Developer setup
A real software project. Use it via its install path; don't expect to read it all.
Runs on
TypeScript
Needs API keys (.env)
TypeScript 52%Python 47%
Where to start reading
  1. 1
    README.md
    Start here — what it does and how to install it
  2. 2
    AGENTS.md
    The instructions the AI actually follows
  3. 3
    .env.example
    The API keys and settings you must provide
  4. 4
    examples/data/qdrant.json
    A worked example — copy this to get going
What's in each folder
plugins/Sub-packages / plugins9 files
examples/Examples you can copy388 files
docs/Documentation327 files
python/Folder3945 files
typescript/Folder3595 files
integ/Folder1777 files
conformance/Folder24 files
scripts/Helper scripts22 files
READMEHas testsDocumentedCI checksExamples includedApache-2.0 licenseUpdated this month
Quick Actions
Details
Creator
strukto-ai
Language
TypeScript
Category
ai-agent
Published
5/6/2026

Are you the creator of this tool? and earn 85% of every sale.

Show it off in your README:

Featured on WorkflowStacks
[![Featured on WorkflowStacks](https://workflowstacks.com/api/badge/mirage.svg)](https://workflowstacks.com/skills/mirage?utm_source=github&utm_medium=badge)
🔥 Hot this week, in your inbox
Skills like this, every Monday.

The five fastest-growing open-source AI skills, ranked by GitHub star growth. One email a week, unsubscribe anytime.

Do this with a working workflow

Pages we wrote and tested that put a skill like this to work — importable n8n templates, outcome guides and Claude Desktop configs.