What we've shipped
Open source. Self-hosted. No accounts required.
Local-first, security-first, modular AI tooling — composable parts for people who want to look beneath the surface.
# Local-first memory across all your coding agents. No API key.
$ curl -fsSL https://fantazm.ai/recall/install.sh | sh
$ recall download-model
$ recall doctor
database ~/.local/share/recall/db.sqlite
embedder all-MiniLM-L6-v2 · 98.9% R@5, on-device
wired Claude Code · Codex · Cursor · Hermes
# open any tool — your context follows you:
› what did we decide about the auth migration?
→ recall_search: per session 56abb5da — "JWT, 15-min expiry"A short list of things we won't compromise on.
Your data, your machine. Nothing leaves the device unless you explicitly opt in. Even our cloud features (when we add them) will be optional, not the default.
Hash-chained audit log, encrypted credential vault, egress proxy, sandboxed skills. Not features — load-bearing structure. If it's not auditable, we don't ship it.
Most agent platforms ship a control plane. We ship a single static binary. If you can run `curl | sh`, you can run our stack.
Claude, OpenAI, Ollama — your choice per session. The agent loop is provider-agnostic. Switch in one line. Bring your own keys.
Each tool we ship is a standalone binary with a wire-protocol API. Use one. Use all. Swap any of them for someone else's. Standard formats (OTLP, OAuth, OIDC) so nothing's a dead end.
Permissive licenses, every commit public, no closed-source core. We'd rather ship a smaller thing that you can fork than a bigger thing you can't see. Issues and PRs are how we ship the roadmap.
Open source. Self-hosted. No accounts required.
In active development.
The agent runtime.
One Go binary. Multi-LLM (13 providers). Encrypted vault, sandboxed skills, hash-chained audit log, scheduled tasks, native Mac menu-bar app + iOS + Android.
Open source. Local-first. No accounts. Pick what you need — there's no platform to sign up for.
