Spooling
Resources
Back to spooling.ai

Security

How Spooling and Spooling Cloud handle your data, what is not in place yet, and how to report a vulnerability.

Reporting a vulnerability

Email security@spooling.ai with the details. Please don't open a public GitHub issue for a security problem. Give us a reasonable chance to fix it before sharing it, don't access other people's data, and don't degrade the service. We will acknowledge your report and keep you updated.

The same policy is on the Security page and in security.txt. What we store and who else touches it is in the Privacy policy.

OSS Spooling

  • Runs entirely on your laptop. The default install does not phone home.
  • Local database on your machine; embeddings come from a local sentence-transformers model, no API key required.
  • MIT licensed.

Spooling Cloud

  • Tenant isolation in the database. Every table that holds session data uses Postgres row-level security, forced even for the application's own database role, so one workspace cannot read another's rows even if a query is wrong. The application role cannot read password hashes or reset tokens; those sit behind a separate role.
  • API keys are stored as SHA-256 hashes, are tied to one workspace, and stop working when their owner leaves it. The cloud never has the raw key after creation.
  • Passwords are hashed with scrypt. Reset, confirmation and email-change links are stored as hashes and are single use. Sign-in and similar endpoints are rate limited.
  • Sessions use a secure, HttpOnly, same-site cookie. "Sign out everywhere" ends every session.
  • CLI sign-in uses a device flow: the key is generated on your machine and the server only receives its hash.
  • Input limits. 256 KB per message, 5,000 messages per session, 100 sessions per batch, request-size caps, and rate limits per key or session.
  • Transport and exposure. TLS through Cloudflare; the server opens no inbound ports other than SSH.
  • Backups are taken nightly and encrypted before they leave the database server. Deleted data can remain in them for up to 21 days.
  • Your data, your control. Download everything or delete your account from Settings.

What is not in place yet

  • No SOC 2 or ISO 27001 report, and no third-party penetration test yet.
  • No two-factor authentication, SSO/SAML or SCIM yet.
  • One server in one region, with no uptime guarantee. Nightly backups mean a failure could lose up to a day of data.
  • Secret redaction happens in the CLI only, is pattern-based and best effort, and can be turned off with --no-redact.

Self-hosted

Self-hosted Spooling Cloud is on the roadmap but not available. Until then the OSS engine is fully self-contained for individual developers, and Spooling Cloud runs on our infrastructure only.

On this page