Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.mirage.strukto.ai/llms.txt

Use this file to discover all available pages before exploring further.

​
How To Read This Matrix

Use this page to pick the first resource to try. If you want the fastest path, start with RAM or disk. If you need real integrations, jump from the setup guide to the resource docs. The Mount Mode column shows which MountMode values the resource supports (read, write, and exec, the last makes a mount executable so commands can launch binaries from it; typically used with Disk or RAM). The Docs column shows where each resource is available, with one icon per supported runtime: Python, TypeScript (Node), and TypeScript (Browser). Click an icon to jump to that runtime’s docs. Sections below mirror the Setup sidebar so you can pick a category and follow the same path through credentials → resource docs.

​
Infrastructure

No external setup, these run locally or against a connection string you already have.
ResourceMount ModeDocsNotes
RAMread, write, exec Best first-run option
Diskread, write, exec Local filesystem bridge
OPFSread, write, execBrowser-only persistent filesystem
Redisread, write, exec Persistent cache-backed workspace
SSHread, write Remote filesystem access

​
Object Storage

ResourceMount ModeSetupDocsNotes
S3read, writeS3 Common cloud object store
R2read, writeR2 S3-style API
GCSread, writeGCS GCP object storage
OCIread, writeOCI Oracle object storage
Supabaseread, writeSupabase Storage-oriented workflows

​
Google Workspace

ResourceMount ModeSetupDocsNotes
Gmailread, partial writeGoogle Mailbox access
Driveread, partial writeGoogle File tree and metadata
Docsread, partial writeGoogle Document content
Sheetsread, partial writeGoogle Spreadsheet data
Slidesread, partial writeGoogle Presentation content

​
Cloud Files

ResourceMount ModeSetupDocsNotes
DropboxreadDropbox OAuth2 + PKCE; Node and browser
BoxreadBox OAuth2 + PKCE or developer token; .boxnote.json / .boxcanvas.json / .gdoc.json decoded

​
Code & DevOps

ResourceMount ModeSetupDocsNotes
GitHubreadGitHub Repository browsing
GitHub CIreadGitHub CI Runs, logs, artifacts
LinearreadLinear Issues and projects
LangfusereadLangfuse Trace exploration

​
Messaging

ResourceMount ModeSetupDocsNotes
SlackreadSlack Channels, messages, files
DiscordreadDiscord Guild and channel history
TelegramreadTelegramChat and file access
Emailread, writeEmail Mailbox workflows; browser blocked by raw TCP requirement

​
Database

ResourceMount ModeSetupDocsNotes
MongoDBreadMongoDB Collection-backed views
PostgresreadPostgres SQL tables exposed as paths

​
Notes

ResourceMount ModeSetupDocsNotes
Notionread, writeNotion Pages and blocks; TS support is browser-only

​
Others

ResourceMount ModeSetupDocsNotes
TrelloreadTrello Boards and cards
PaperclipreadPaperclipResearch workflows
Semantic ScholarreadSemantic Scholar Paper search and metadata
PostHogread Projects, insights, recent events
Vercelread Projects, deployments, build logs

​
Agent Frameworks

Mirage drops into the major agent application frameworks. Each adapter exposes a Workspace as the agent’s filesystem and shell. Click an icon to jump to the integration docs.
FrameworkDocsNotes
OpenAI Agents SDK MirageShell and MirageEditor plug into shellTool and applyPatchTool.
Vercel AI SDK mirageTools() returns five typed tools for generateText / streamText.
LangChain (deepagents) LangchainWorkspace backend for deepagents. Node only.
Pi Coding AgentMirage extension for @mariozechner/pi-coding-agent. Node only.
MastramirageTools() for Mastra Agent definitions. Node only.
Pydantic AIFor pydantic-ai and pydantic-deepagents.
CAMEL-AIFor CAMEL ChatAgent.
OpenHandsFor the OpenHands agent SDK.
Claude Code (CLI) Mount via FUSE; run claude against the mountpoint.
Codex (CLI) Mount via FUSE; run codex against the mountpoint.
  • Use RAM if you want to learn Mirage itself.
  • Use Disk if you want local file interoperability.
  • Use S3 or GitHub if you want a realistic first external integration.