Introduction
What Spooling and Spooling Cloud are, and how they fit together.
Spooling solves agentic amnesia: the moment a developer switches AI coding tools, none of yesterday's reasoning, fixes, or dead ends crosses with them. Memory lives at the seams between tools.
There are two products. The same CLI talks to both.
The OSS engine: Spooling
spooling is the open source Python CLI and local indexer. It runs on your laptop, parses sessions from the AI coding tools you use (Claude Code, Cursor, Codex, Copilot, Windsurf, Kiro, Antigravity, Gemini Code Assist, OpenCode and Cortex Code), and indexes them in a local database on your machine.
- MIT licensed
pipx install spooling: no Docker, Postgres or Node required- Local sentence-transformers embeddings, no API key required
- Searchable from the CLI, a local web UI, or via MCP
Get started with the OSS engine →
The hosted workspace: Spooling Cloud
Spooling Cloud is the multi-tenant hosted product at app.spooling.ai. The OSS engine on each developer's laptop pushes sessions to a shared workspace. It is free while in early access. The team gets:
- Personal and team workspaces, with invite links and member roles
- Cross-teammate semantic and keyword search
- Per-person and per-project views, with usage and cost rolled up by provider
- An MCP endpoint your agents can call as a tool
- A REST API
- Sign in with Google, GitHub or email, plus data export and account deletion from Settings
Not available yet: SSO/SAML, SCIM, audit-log export, compliance certifications and self-hosting. They are on the roadmap for larger teams.
How the pieces fit
┌────────────────────┐ ┌──────────────────┐
│ Developer laptop │ │ Spooling Cloud │
│ ─────────────── │ spooling push │ api.spooling.ai │
│ OSS spooling CLI │ ───────────────► │ multi-tenant │
│ Claude / Cursor / │ │ Postgres │
│ Codex / ... │ │ │
└────────────────────┘ └────────┬─────────┘
│
┌──────────────────────┼─────────────────────┐
▼ ▼ ▼
┌──────────────┐ ┌──────────────┐ ┌──────────────┐
│ Web app │ │ MCP endpoint │ │ REST API │
│ app.spooling │ │ for agents │ │ for scripts │
└──────────────┘ └──────────────┘ └──────────────┘Every developer installs the same OSS spooling binary. spooling login is what connects it to a workspace.
What's in these docs
- Get Started: install, log in, push your first session
- CLI Reference: every command, every flag
- Cloud: workspaces, cross-workspace patterns, REST + MCP API
- Providers: what we parse from each AI tool, where it lives on disk
- Security and privacy: what we store, who can see it, how to delete it