Persistent, private memory for Claude Code — that never spends your tokens on itself.
Every session is archived into a local SQLite database and made searchable: last month's work is one subrosa search away, relevant past sessions resurface when you type a related prompt, and Claude stops rediscovering what it already figured out. Pulling up an old answer costs a few hundred tokens; having Claude re-derive it from scratch runs into the thousands.
- No LLM calls to save memory. Saving a session is plain-text parsing — zero tokens. Most memory plugins run your sessions through an LLM to save them (the comparison has the numbers, from their own docs).
- Hard token limits, set in the code. Recall adds ~180 tokens on a strong match, usually nothing — it stays silent otherwise. The always-loaded index is capped at 23 KB. Check it yourself, or see where your tokens go.
- One 3.7 MB static binary. No daemon, no worker port, no background process — a hook fires, finishes in under 10 ms, and exits.
- Your transcripts stay on your machine. The binary makes zero network calls — no cloud, no telemetry — and obvious secret shapes are masked before storage. Verify every claim yourself.
Why "subrosa"? ·
Quick start ·
What it does ·
Commands ·
The memory workflow ·
Make Claude search the archive itself
Where your data lives ·
Privacy model ·
Proof ·
Performance ·
Development ·
How it compares ·
FAQ
Why "subrosa"?
In ancient Rome, a rose over the table meant everything said under it stayed in the room — sub rosa. That's the contract here: every transcript stays on your machine, in a local database, secret shapes masked before storage. No cloud, no telemetry.
Quick start
Inside Claude Code, run these two commands:
/plugin marketplace add ij5a/subrosa
/plugin install subrosa@subrosa
Then start a new Claude Code session (quit and reopen, or run claude again). That's the whole install:
- On first start, the plugin downloads the right prebuilt program for your computer (~2 MB, checksum-verified against this repo) and archives every session already on your disk.
- After that it runs itself: archives each session as you work and again when it ends, shows Claude related past sessions when you prompt, and notes when ended sessions are waiting to be saved into long-term memory.
The one-time download fetches the program only — your data never moves.
One more command, for the best experience — it teaches Claude to search the archive at the start of any task instead of waiting for you to ask (what it adds; safe to re-run, ~150 tokens):
~/.claude/subrosa/bin/subrosa init --claude-md # or just: subrosa init --claude-md, with the CLI below
Optional: the subrosa command in your terminal
The plugin works without this. Install the CLI to also search and manage the archive yourself:
# homebrew
brew install ij5a/tap/subrosa
# or the install script (prebuilt, checksum-verified)
curl -fsSL https://raw.githubusercontent.com/ij5a/subrosa/main/install.sh | sh
# or from source, if you have Rust
cargo install --git https://github.com/ij5a/subrosa
The plugin finds a binary on PATH and uses it, so both stay in sync. Two useful first commands:
subrosa setup # one-time: pick where backup snapshots mirror to (iCloud / Dropbox / ... / none)
subrosa # the dashboard
What it does
