Agenten stärken
Installation vom Herausgeber. Kein PlugMCP-Endpunkt.
Host a copy with our keys and logs.
Noch kein paketiertes Release in der Registry. Folgen Sie der README des Repositorys.
Remote-URL oder ein Befehl auf Ihrem Rechner? Worin sich die beiden Installationen unterscheiden
Auf PlugMCP ausführen
A new server you create. It is not this listing's official endpoint. Use the publisher install when you want their URL and their credentials.
StartenVom Herausgeber
Session memory for Claude Code — a second brain that builds itself.
Every Claude Code session — across every project, on every machine — is captured into a plain Obsidian markdown vault you fully own. Zero config, no LLM of its own, no cloud. SuperBrain observes the sessions you're already running and writes the journal you never have time to.
Session memory for Claude Code. Auto-captures every session into a plain Obsidian markdown vault. Zero config. MIT. macOS + Linux.
Status: in active development. Interfaces and behavior may change before a tagged release.
Every Claude Code session ends and the reasoning vanishes. Decisions, trade-offs, the "we tried X and it didn't work" history — gone with the terminal scrollback. CLAUDE.md fixes the rules; SuperBrain fixes the journal. The Claude Code memory ecosystem has split in two, and neither half is what you actually want:
No mature tool does all of: globally installed → automatic capture → into a plain Obsidian vault → with incremental daily notes → that you fully own and can git-sync. SuperBrain is that missing bridge.
This isn't vibes: ~23 prior-art projects and the current Claude Code platform were surveyed, and every architectural decision was challenged before a line was written.
# In Claude Code:
/plugin marketplace add m3talux/superbrain
/plugin install superbrain
That's it. On the first session the plugin runs a one-time setup (installs its
search dependencies in the background) and tells you it's doing so; capture is
fully active from the next session. SuperBrain writes to its own vault at
~/.superbrain/vault — a fixed, predictable location with no environment
variables to set. Point Obsidian at that folder and you're done.
If you already have an Obsidian vault you want to bring along, see
Vault setup below — /superbrain:migrate is the preferred path.
Installed at user scope, the plugin's hooks register for every project automatically — there is nothing else to do, ever.
If SuperBrain saves you context, please star ⭐ — it's the only signal we have.
Tested on Node 20 LTS and Node 22.
flowchart LR
A[Tool use / prompt] -->|PostToolUse, UserPromptSubmit<br/>async, no LLM| B[NDJSON log<br/>+ salient markers]
B --> C{Checkpoint?<br/>PreCompact / SessionEnd /<br/>Stop if pending}
C -->|detached, lock-serialized| D[sb-distill<br/>claude -p]
D --> E[Router]
E --> F[(Obsidian vault<br/>notes only)]
D -.->|incremental rebuild| I[Daily note for today]
I --> F
C -.->|byte cursor| B
PostToolUse / UserPromptSubmit hooks append a compact event line to a per-session NDJSON log. No LLM on this path, so it can never rate-limit, stall, or disrupt your turn.PreCompact, SessionEnd, or a pending-gated Stop, one detached, lock-serialized claude -p reads the delta since a byte-offset cursor, classifies it, and writes routed notes via an in-process vault writer.daily/<date>.md from the per-session state (digest line + routed-note links + threads). Deterministic concat, no extra LLM call.Capture
projects/<slug>.md (stack / architecture / top-level folders / key files / docs / conventions / open questions). Once per project, ever. Never overwrites an existing note.git-portable, zero lock-in.trash/Search & recall
superbrain_search MCP)superbrain-recall skill + stdio MCP server (superbrain_search) for model-invoked deep searchPersonalization & journaling
CLAUDE.md)Planned
maps/) + a lint pass./superbrain:injectAuto-capture handles everything SuperBrain can observe inside a Claude session. When you want to record something it didn't see — a random side thought, a meeting summary from elsewhere, a note imported from another system — use the inject command:
/superbrain:inject I just realized the auth-config endpoint probably needs a TTL field
/superbrain:inject --from-file ~/Downloads/meeting-summary.md
/superbrain:inject --project wcloud "transport is HTTP, not gRPC"
What it does:
capture/. No LLM call.decisions/, project facts to the matching projects/<slug>.md, references to people to people/, everything else to capture/.source: inject provenance so it's trivially auditable.Safety rails: inject never creates new project notes (use /superbrain:discover for that), never reshapes preferences, and never silently loses your input — if the distiller returns nothing, your text is written verbatim to capture/.
~/.superbrain/vault/
├── projects/ one note per project — status, decisions, current focus
├── people/ one note per person — role, context, threads
├── decisions/ atomic, date-prefixed ADR-style notes
├── daily/ auto-written daily activity
├── lessons/ durable, generalizable rules learned from your pushback
├── capture/ raw inbound; triage tag for `/superbrain:inject` items
├── sessions/ per-session flight recorder — raw turn log + checkpoint digest (pruned after 30 days)
├── meta/ preferences.md — deduplicated profile auto-injected at SessionStart
├── maps/ auto-generated Maps-of-Content (planned)
└── index.md catalog — the primary navigation surface
System telemetry (NOT in the vault — lives next to it in ~/.superbrain/):
~/.superbrain/
├── sessions/ per-session NDJSON event streams + cursors
├── transcripts/ full session transcripts (when captured)
└── index.db hybrid search index (FTS5 + sqlite-vec)
Generated files are namespaced (daily/, index.md) so the auto-rebuilders never touch notes you authored.
SuperBrain's vault lives at ~/.superbrain/vault — fixed, predictable, no env vars. Two ways to bring an existing Obsidian vault into the picture:
/superbrain:migrate # auto-detects your vault
/superbrain:migrate ~/Documents/MyVault # explicit path
/superbrain:migrate --dry-run # preview without writing
/superbrain:migrate reads your existing Obsidian vault (the source is never modified) and copies its content into ~/.superbrain/vault, fitting it into SuperBrain's structure (projects/, decisions/, people/, daily/, etc.). From then on, SuperBrain owns and extends a vault it fully understands — its router, daily notes, and index all behave predictably. This is the recommended path for almost every user.
/superbrain:adopt ~/Documents/MyEstablishedVault
/superbrain:adopt does not copy anything; it records your existing path as the vault location and starts writing into it directly. Use this only when you have a heavily established Obsidian setup whose structure you do not want to change — broken into folders, plugins, layouts, or conventions SuperBrain isn't designed around.
The trade-off: SuperBrain will work on top of whatever's already there rather than alongside a structure it owns. The router may write into folders that don't match your existing convention; auto-generated daily notes will live alongside your hand-authored notes; the search-index reconciler will still self-heal but has more drift to manage. Migration sidesteps all of that.
When in doubt: migrate, don't adopt.
SuperBrain is configured by command, not by environment. There are no required env vars. The single optional one:
Heads-up (2026-06-15): background
claude -p/ Agent-SDK usage on subscription plans draws from a separate capped monthly credit after this date. If captures stop, SuperBrain surfaces a one-time notice on session start — setANTHROPIC_API_KEYto switch to the API path.
SuperBrain deliberately does not push your vault anywhere. It writes plain markdown into a local folder and nothing leaves the machine. This is a design choice, not a missing feature: a network operation in the capture path could hang, stall on an auth prompt, or fail — exactly the failure mode the "never disrupt the session" guarantee exists to prevent. Backup and multi-machine sync are a solved problem, and dedicated tools do it better and safer than a hook can.
What SuperBrain does guarantee is that it composes cleanly with whatever sync
you choose: writes are append-or-create and idempotent, and the search index
self-heals on session start after the vault changes underneath it (an
external git pull, an Obsidian edit, a Syncthing update). So you can layer any
of the following on top with no special configuration:
~/.superbrain/vault)? Pick one:
Make it a git repo and use Obsidian Git for scheduled commit/push.
Point Syncthing or a file-sync tool at the folder.
Or a DIY scheduled job — e.g. a cron / launchd entry running, say every
15 minutes:
cd ~/.superbrain/vault && git add -A \
&& git commit -q -m "vault sync $(date -u +%FT%TZ)" \
&& git pull --rebase --autostash && git push
Whichever you choose, treat it as backup / sync owned by you, not by the
plugin. Running the same vault from two machines concurrently can still produce
git conflicts in regenerated files (index.md, today's daily/) — Obsidian
Git's conflict UX or a single-writer-at-a-time habit avoids that. SuperBrain's
job is to stay idempotent and drift-tolerant so any of these Just Work; it is
not to own your remote.
SuperBrain runs a one-time npm ci + npm rebuild better-sqlite3 on first session start. If ~/.superbrain/last-failure.txt reports a bootstrap failure, the message includes a platform-specific hint. Common cases:
npm ci failed — network / proxy issue. Retry. If you're behind a corporate proxy, configure npm config set proxy ... and retry.sudo apt install build-essential python3 and retry.xcode-select --install and retry.SuperBrain re-tries bootstrap on every session start, so once the underlying issue is fixed, the next session picks it up automatically.
Enforced, tested invariants — not aspirations:
PreCompact never blocks compaction; a crashing hook is impossible by construction..trash/.SessionStart.npm ci
npm run typecheck # tsc --noEmit, zero errors
npm run build # → dist/
npm test # full suite (unit + integration + fresh-clone E2E)
npm run release:check # build is reproducible: committed dist/ matches source
Contributions welcome — see CONTRIBUTING.md.
Standing on the shoulders of: Andrej Karpathy's LLM Wiki pattern, claude-mem (global-plugin distribution + context injection), basic-memory (Obsidian-native markdown memory), obsidian-second-brain (AI-first note rules), and claude-memory-compiler (session-triggered rollups).
MIT © m3talux
Session memory for Claude Code. Auto-captures every session into a plain Obsidian markdown vault. Zero config. MIT. macOS + Linux.
Tools
Noch keine Tool-Liste. Remote-Server erhalten eine nach dem ersten Health-Check.
FAQ
Kopieren Sie auf dieser Seite die stdio-Konfiguration für Cursor, Claude Desktop, Claude Code oder VS Code. Sie nutzt die Installation des Herausgebers, keinen PlugMCP-Proxy.
In diesem Eintrag ist keine erforderliche Umgebungsvariable aufgeführt.
Superbrain läuft lokal über stdio.
Zuletzt geprüft: 2026-10-27. Session memory for Claude Code. Auto-captures every session into a plain Obsidian markdown vault. Zero config. MIT. macOS + Linux.
