Security fixes are applied to the latest release line only; older minor lines do not receive backports.

Please do not report security vulnerabilities through public GitHub issues, discussions, or pull requests.

Instead, email evermind@shanda.com with:

• A description of the vulnerability and its potential impact
• Steps to reproduce, or a proof-of-concept
• The affected version / commit
• Any suggested mitigation, if you have one

We will acknowledge your report within 5 business days, keep you informed of progress, and aim to ship a fix or mitigation before any public disclosure. Reporters are credited in the advisory and the release notes unless you prefer to remain anonymous.

Confirmed issues are published as GitHub Security Advisories, each carrying the affected version ranges and the release that fixes them:

https://github.com/EverMind-AI/EverOS/security/advisories

To check whether an installation is affected by a past issue, read the version range on the advisory itself — the range is authoritative.

EverOS runs as a local-first service for single users or small teams (Markdown + SQLite + LanceDB on the local filesystem). Please keep the following in mind:

• Exposing the HTTP API (everos server) to an untrusted network is outside the supported threat model — it assumes a trusted local caller. The server binds to 127.0.0.1 by default (env EVEROS_API__HOST) so a fresh install is loopback-only. Only set the bind to 0.0.0.0 (or any routable interface) after you have placed your own gateway / auth layer in front; everos server start will log a warning when you bind to 0.0.0.0.
• Documents you ingest are untrusted input. Filenames, metadata, and content that originate outside your control are parsed, indexed, and written to disk. An instance that ingests third-party documents is processing untrusted data even when the API itself is reachable only from loopback.
• Secrets (LLM / embedding API keys) live in your local .env; protect that file as you would any credential. EverOS never transmits them anywhere except the providers you configure.
• Memory content is stored as plaintext .md files; apply OS-level file permissions or disk encryption if your data is sensitive.