Compartment

Superior agentic memory, encrypted at rest.

Your AI agent forgets you the moment the session ends. Compartment ends that. With Compartment, your AI agent gets better with experience: it keeps every decision, preference and detail you give it, permanently, encrypted, on your own computer. Hermes, Claude, OpenClaw and other AI Agents can install in one command. One fully-transferable memory store is shared simultaneously by all agents on the computer. 100% offline: no network, no API key, no cloud account, no telemetry. The embedding model ships inside the package, and a full search returns in about 12 ms, beating the round-trip a hosted memory charges you for. Every byte at rest is AEAD-encrypted, the embedding vectors included, and only your passphrase opens it.

Unlike other agentic memory, we offer an option to start off with memory - 6,718 curated facts seeded at install: the physical constants and unit conversions, 800+ hardware facts with real specs (Apple silicon, PCs, CPUs and GPUs, phones, game consoles, Raspberry Pi, storage, displays, connectors) to provide a map of computer geography, operating system versions and release names, network ports and HTTP, file signatures, character encodings, shell and Unix internals, git, regex, SQL and hashing, ISO country, currency and time codes, and more. Allows an offline agent to operate better without internet, and an online agent to operate faster and more accurately.

More secure, by construction. Every byte at rest is authenticated-encrypted, the embedding vectors included (most tools leave those in the clear, and vectors can be inverted back toward text). Deletion is cryptographic: destroy the record's key and it is gone, unrecoverable. Tampering is detected, history is hash-chained, and the vault locks itself on restart or power loss. It runs fully offline: a runtime guard aborts on any network attempt, and CI proves it on three operating systems.

Not one step harder. One command installs it, creates the vault, and wires your agent. No API key, no cloud account, no daemon. Unlock when you want to use it; lock when you want it closed. By default an unlock stays open for weeks (until restart or you lock it), like any app you leave running. The security is free at the point of use because it falls out of the architecture, not out of your patience: keeping plaintext off disk forces the index into RAM, and a RAM-resident index is also the fastest one there is. Secure and fast are the same choice here, and neither costs you a configuration step.

Install

One line install, works on all operating systems.

pip install compartment && compartment init

Then connect it to the agent you use:

compartment integrate claude

claude, hermes and openclaw are the three auto-connect targets. Each one also gets the /compartmentalize skill installed into its own skills directory.

/compartmentalize saves the conversation before it is thrown away. Every agent eventually compacts or summarizes a long session, and the summary is written by a pass that has no tools, so nothing can be stored from inside it: whatever the model did not think to save is simply gone. Type /compartmentalize and the whole conversation is swept into the vault first - people and contacts, credentials and where they live, URLs and hosts, decisions and the reasoning behind them, and a narrative of the session itself. Then compact, and nothing is lost. It works on its own at any point too.

One click install (for people not good with command line). Download Compartment.pkg from the latest release and open it. Python, the embedding model and every dependency are inside it. macOS only.

After install, everything is managed from the app: the menu bar on macOS, the notification area on Windows, and a window from your applications menu on Linux.

Compartment is an MCP server, so it works with all MCP capable agentic AI out of the box. Every option is in Configuration.

What sets it apart

Install it in one step

• One command installs it, creates the vault, and wires your agent. No API key, no cloud account, no daemon.
• On the Mac, open one .pkg and you are done. Python, the embedding model and every dependency are inside it.
• An app runs the whole thing without a terminal on all three systems: vault state, unlock, lock, and the last five memories it saved. The macOS menu bar, the Windows notification area, a window on Linux.
• Every feature toggles in that panel instead of a config file: model-independent capture, starter facts in search, auto-lock.
• /compartmentalize is installed into every agent it connects, so one command banks a whole conversation before compaction throws it away.
• Your vault ships full. The 6,718 seeded facts are ordinary memories, editable and forgettable, and one switch keeps them out of search.
• Runs under what you already use: Hermes ("no setup needed"), Claude Code and Desktop over MCP, OpenClaw, every MCP client, plus a CLI for scripts and cron.

Remembers the right things

• "OK" is a decision, and Compartment files it as one, with the question it answered. That is the record you need later.
• Decisions beat preferences, preferences beat machine details, machine details beat chatter. A fixed ranking, not a model's mood.
• It forgets nothing. Small talk is kept and ranked last.
• It replaces your host's built-in memory instead of fighting it: imports what Claude Code already wrote, then supersedes it.
• It captures even when the model does not cooperate. A hook writes the fact whether or not the model calls the tool.
• A graph, not a pile. Explicit relations with validity windows answer who worked where, and when.

Search that beats a network call

• 0.68 ms vector search. About 12 ms for the full hybrid pipeline. A cloud memory spends longer than that saying hello.
• Exact below 20k records: recall = 1.0 by construction, not an approximation.
• Hybrid always: meaning and keywords, fused.
• One pinned embedding space, enforced every time the vault opens, so your comparisons stay valid forever.

Encrypted, offline, and yours

• Every byte at rest is AEAD-encrypted, embedding vectors included. Most tools leave vectors in the clear, and vectors invert back toward text.
• Only your passphrase opens it. Compartment generates no password, no seed, no recovery phrase, and holds no credential you do not.
• Add a keyfile and unlock takes two factors. Both feed Argon2id together, so it is arithmetic, not a policy check.
• forget --shred destroys the record's key. The content is mathematically unrecoverable, not marked deleted.
• Restart or power loss locks it, and the agent has a panic lock that clears every credential instantly.
• 100% offline. A runtime guard aborts on any network attempt, and CI proves it on Linux, macOS and Windows. Zero open ports. No telemetry, ever.
• No LLM inside. Embeddings run locally in under 300 MB, and judgment stays with the model you already pay for.
• Tamper-evident: hash-chained audit log, sealed journal, verified kill-9 crash recovery.
• One portable file. Move a locked vault anywhere, and lock --sign seals it with an Ed25519 manifest anyone can verify without a credential.
• compartment dash puts the entire vault on a local page: 127.0.0.1 only, random token, read-only.

The memory logic

Full write-path, decision math, and comparisons in docs/MEMORY.md. The load-bearing ideas:

Nearly everything is stored; nothing important is buried. Only empty turns are dropped. A bare "OK" is not noise, it is a decision: when the agent asks "Want me to send this reply to the client now?" and the user answers "OK", Compartment resolves the question from the conversation and stores [decision 2026-07-20] Approved (answered "OK"): Want me to send this reply to the client now? at the top importance tier. Asking "did the user say to email the client?" later retrieves exactly that record.

Deterministic importance tiers rank recall: decisions/consent 0.90, personal facts and preferences 0.80, the user's machine and configuration 0.75, other substantive statements 0.55, pleasantries 0.20 (kept, ranked last). Importance multiplies a match rather than adding to it, so it settles near-ties in favour of what matters and can never surface a memory for a question it has nothing to do with. The whole scoring model, and the numbers it was chosen against, are in The mathematics. The agent learns the user and the computer first, the world second, and forgets nothing.

One memory, not two. Agent hosts increasingly ship a memory of their own - Claude Code keeps per-project Markdown files with an auto-loaded index. Two memories means facts land in whichever one the model happened to think of, and neither is complete. Compartment takes over on install: it imports what the file memory already holds, and both the MCP handshake and the managed CLAUDE.md block tell the model that Compartment supersedes it - write every new memory here, treat the files as a read-only archive. One vault, encrypted, shared by every agent and project on the machine. Nothing is deleted; the files stay exactly where they were.

Capture that does not depend on the model. Instructions are a request, and a host that declares its own memory in its system prompt outranks anything a tool says. So compartment integrate claude also installs a PostToolUse hook: when Claude Code writes a memory file, the fact lands in the vault whether or not the model ever thought about Compartment. The hook is additive and idempotent (your other hooks are untouched, settings.json is backed up first), it exits successfully no matter what - a memory tool must never break your editor - and it stays quiet when the vault is locked. compartment hook status | install | uninstall, or integrate claude --no-hooks.

An app on all three systems, from the same one install. pip install compartment && compartment init sets it up on macOS, Windows and Linux. There is no separate package, no extra to remember and no second command. On the Mac you can instead open Compartment.pkg from the latest release, which carries Python and every dependency inside it, so there is nothing to install first. compartment init --no-app skips the app for headless boxes and CI.

The same panel, in the place each system keeps things like this: the menu bar on macOS, the notification area on Windows, and on Linux an ordinary window, with Compartment in your applications menu. Linux gets a window rather than an icon deliberately. Whether a tray icon appears there depends on the desktop, and on GNOME or Wayland it can simply never show up with nothing said, which is the worst way for the control that unlocks your memories to fail.

The panel shows whether the vault is open, how much it has learned, the three settings worth changing day to day (capture hook, whether starter facts join searches, auto-lock), which agents are connected and buttons to connect them, and the last five things it remembered. Unlock, lock and change your passphrase there too, without opening a terminal. It holds no vault in memory - state comes from the CLI, so an idle app costs nothing.

See what it just learned. compartment recent lists the newest memories, newest last, hiding the thousands of seeded starting facts so the handful that real use produced are actually visible - and compartment status reports organic_records beside the total, so a vault that has learned nothing can never look busy. Same view over MCP as memory_recent.

One pinned embedding space. The model's SHA-256 is recorded in the vault and enforced at open; cosine comparisons stay mathematically valid forever instead of silently degrading when a model changes. Migration is explicit: compartment reindex --re-embed.

No LLM inside. Embeddings run locally (bundled 384-dim int8 ONNX model, <300 MB RAM). Judgment belongs to the host model you already run, via memory_store / memory_forget; Compartment contributes deterministic capture, encryption, and total recall. That split is what makes the offline guarantee absolute and every decision reproducible. Pair Compartment with an offline LLM and the whole agent stack can run usefully with no network at all.

The mathematics

Everything below lives in one file, src/compartment/ranking.py, which the vault, the dashboard and the benchmark all import. A benchmark score is therefore a measurement of the product and not of a copy of it that has drifted.

Storage: a memory is embedded in windows, not truncated

The encoder reads 512 tokens. Text past that is not weighted less, it is not seen at all, so a long memory used to be searchable only by its opening. On a real 6,705-memory vault, 40% of records ran past the window and 57.6% of the whole corpus was invisible to semantic search.

So a record is embedded as overlapping windows of W = 448 tokens at a stride of S = 384, giving 64 tokens of overlap so no fact is cut in half by a boundary, and the record is scored by its best window:

windows(d) = ceil( max(0, tokens(d) - W) / S ) + 1        capped at 64

s_vec(d)   = max over windows w of d :  cos(q, w)

Max-pooling, not averaging: a memory is relevant if any part of it is, and an average would punish a long memory for the parts that are about something else. With one window per record it reduces exactly to the old behaviour, so it can never be worse for a short memory. The cost is small because most memories are short: on that vault, 6,705 records produced 6,785 windows.

Windows are measured in model tokens, never characters. A character budget is wrong by a factor of three between prose and a hex digest, and being wrong here means silently dropping the end of a memory.

Recall: two channels, combined as evidence rather than added

Two indexes look for a memory and they answer different questions. The vector index answers what does this mean. The keyword index answers what does this say. Their scores are not denominated in the same thing, and combining them is the entire difficulty.

The obvious move, and what Compartment shipped until now, is to add them. Adding is the wrong operation: it lets a merely-good semantic match outvote conclusive literal evidence. Searching a real vault for a commit sha occurring in exactly one memory out of 6,705 returned that memory below ten paraphrases of it - the keyword index had ranked it first and the sum buried it.

The two channels are not addends, they are alternatives: either one alone can establish relevance. That is a soft OR over independent evidence,

P(relevant) = 1 - (1 - p_vec)(1 - p_lex)

and the score is its logarithm, which ranks identically while continuing to spread results apart near the top instead of saturating at 1:

score(d) = - w_vec · log(1 - p_vec(d))  -  w_lex · log(1 - p_lex(d))

w_vec = 0.75      w_lex = 0.25

Either channel approaching certainty carries the memory on its own, and neither can veto the other.

Reading a cosine as a probability. An L2-normalized encoder gives cosines that are comparable across queries, so they map through fixed bounds. Per-query min-max normalization is the obvious alternative and it is a trap: it rescales the best hit of a hopeless query up to 1.0 and throws that calibration away.

p_vec(d) = clamp( (cos(q, d) - 0.25) / (0.85 - 0.25),  0,  0.88 )

That ceiling of 0.88 is doing real work. A cosine is a similarity, never an identity: an encoder can say this is about the same thing, but it can never say this is the record you named. A literal match on a string unique to one memory can say exactly that. So the semantic channel is capped below the certainty the literal channel may reach, and the bound is forced rather than chosen - the literal channel tops out at 0.25 · -log(1 - 0.999) = 1.727, so the cap must satisfy 0.75 · -log(1 - cap) < 1.727, giving cap < 0.90.

Reading a keyword hit as a probability, and deliberately not with BM25. BM25 answers how well does this match, which is not what settles a contest against a semantic hit. What settles it is how unlikely the match was by chance. So each query term carries its self-information over the vault, and a memory scores the fraction of the query's information it accounts for:

I(t)     = log( N / (1 + df(t)) )                     N = records in the vault

p_lex(d) = ( Σ I(t) for query terms t present in d ) / ( Σ I(t) for all t )

A term unique to one memory is near-conclusive evidence. A term appearing in a tenth of the vault is nearly none, whatever its BM25 happens to be. This is the piece that makes a literal hit and a semantic hit comparable at all.

The keyword index is queried as AND first, since an exact phrase match is the strongest signal available. FTS5's implicit AND means a nine-word question has to appear word for word, so when AND finds nothing it falls back to OR over only the terms carrying information - anything appearing in more than 10% of records is dropped. That ceiling is measured from the vault rather than taken from an English stopword list, so it behaves the same for a vault full of code, of names, or of another language.

A small rank-agreement residue is added, the one thing reciprocal-rank fusion is genuinely good at, sized to break ties rather than decide them:

+ w_rrf · k · [ 1/(k + rank_vec) + 1/(k + rank_lex) ]      w_rrf = 0.10, k = 20

Importance ranking: priors multiply, they never add

final(d) = score(d) · ( 1 + w_imp · (2·importance(d) - 1)
                          + w_rec · 2^( -age_days(d) / 180 ) )

w_imp = 0.15      w_rec = 0.10

Multiplicative, so a prior can only reorder a memory that already matched. An additive prior lets a very important memory surface for a question it has nothing to do with, which is how a memory system starts feeling haunted. A memory that matched nothing scores zero, and nothing can lift it off zero.

Centred on the 0.5 default, which is why 2·importance - 1 appears rather than importance. Every unweighted memory carries 0.5, including the thousands of starting facts a vault ships with. Uncentred, they all collect the same silent boost, which is another way of saying importance did nothing at all. Centred, an unweighted memory is exactly neutral and a deliberate weight is the only thing that moves.

The tiers the capture path writes: decisions and consent 0.90, personal facts and preferences 0.80, the user's machine and configuration 0.75, other substantive statements 0.55, pleasantries 0.20. Recency halves every 180 days.

Retrieval order, and why the pool is wide

Namespace, tag, date and starter-fact filters run after ranking, so a candidate pool sized to the number of results requested can be emptied by them while matching memories sit just past the cut. The pool starts at 200 per channel and widens up to three times when filtering leaves too few.

Measured

Against the previous scorer, end to end through Vault.search, on a real 6,705-memory vault with 44 queries in four families:

	before	after
Recall@1	0.523	0.773
Recall@5	0.705	0.977
MRR@10	0.601	0.845
nDCG@10	0.627	0.878
exact identifiers found in top 5	4/10	10/10
facts past the encoder window	0/6	5/6
paraphrases	16/16	16/16
median search latency	4.4 ms	11.6 ms

Nothing regressed in any family. The weights were chosen from a sensitivity sweep and are deliberately round: the result is flat around them, because a ranker that only works at w_lex = 0.37 is a ranker that does not work.

Wiring each agent

One command per platform. Each installs the package, creates your encrypted vault, and wires the agent.

Every one of these is also a button in the app. Click the Compartment icon in your menu bar or notification area, and under CONNECT AN AGENT press Claude, Hermes or OpenClaw. The button runs the same compartment integrate command for you, so nobody has to open a terminal a second time.

Claude (Code + Desktop) - macOS / Linux:

pip install compartment && compartment init && compartment integrate claude

Windows (PowerShell):

py -m pip install compartment; compartment init; compartment integrate claude

Registers the MCP server with the Claude Code CLI (user scope, all projects), imports any memories Claude Code already wrote to its own file-based memory (copy-only - the Markdown files are never modified; --no-import opts out, compartment import-claude does it later), and prints the Claude Desktop config block. The server describes itself over the MCP handshake - it tells the model to recall before answering and to store durable facts, credentials, names, and decisions - so Claude treats Compartment as its memory with no hand-written instruction; integrate claude also writes a managed, idempotent block into your CLAUDE.md as backup.

Hermes - macOS / Linux:

pip install compartment && compartment init && compartment integrate hermes

Windows (PowerShell):

py -m pip install compartment; compartment init; compartment integrate hermes

Installs the provider plugin, wires the Hermes venv, and runs hermes memory setup compartment. Compartment then appears in the hermes memory setup picker beside hindsight and mem0, the only entry marked "no setup needed": no API key, no cloud account, no daemon. Verify with hermes memory status. See everything Hermes remembers at any time with compartment dash - one command, and the vault opens in your browser (memories by kind, growth, the relation graph, live search); Ctrl-C closes it.

OpenClaw - macOS / Linux:

pip install compartment && compartment init && compartment integrate openclaw

Windows (PowerShell):

py -m pip install compartment; compartment init; compartment integrate openclaw

Writes the mcpServers entry into ~/.openclaw/openclaw.json (with a backup), then: openclaw gateway restart and confirm with openclaw mcp list.

Any MCP client - macOS / Linux / Windows:

pip install compartment && compartment init

Then add the server to your client's MCP config (stdio transport, no API key, no environment variables):

{
  "mcpServers": {
    "compartment": {
      "command": "compartment",
      "args": ["serve"]
    }
  }
}

--vault and --caller are optional (compartment --vault PATH --caller NAME serve); the defaults use ~/.compartment/memory.vault with caller user. Client-by-client walkthroughs in docs/INTEGRATIONS.md.

Measured, on an 8 GB baseline laptop

Every number below is reproducible on your machine with compartment selftest and compartment bench.

Metric	Measured
Fresh install → open vault, offline	seconds, zero network
Vector search, 20k records (HNSW)	p95 0.68 ms
Full hybrid search (embed + windows + BM25 + evidence fusion)	median 11.6 ms, p95 14.7 ms
Peak RSS, model + vault + index resident	319 MB
Store one memory (embed + encrypt + fsync journal)	~40 ms
Wheel size, model included	~30 MB
Test suite (crypto, tamper, crash, offline, concurrency, 2FA, graph, dash, ranking)	566 tests, ~110 s

A single network round-trip to a cloud memory API costs more than this entire pipeline. The property that makes Compartment secure (no plaintext index ever on disk, so all search is RAM-resident) is the same property that makes it fast: below 20k records search is exact SIMD matrix math, recall = 1.0 by construction; above it, SIMD HNSW at ~99% recall.

Agent-native by design

Compartment is built to sit under agents you already use, not as a separate app you babysit.

• Hermes native provider - shows up in hermes memory setup with "no setup needed". Turns sync automatically; search injects only what is relevant, tagged as data not instructions.
• Claude over MCP - one integrate claude step registers the server and gives you the Desktop config block plus a managed CLAUDE.md block so memory is part of normal work.
• OpenClaw and any MCP client - same stdio server, zero open ports, same tools (memory_search, memory_store, memory_forget, lock).
• A memory graph, not just a memory pile - memory_link records explicit relations (who works where, what belongs to what), with optional validity windows; memory_relations answers entity, predicate, and as-of queries. Deterministic storage, host-model judgment - the same split as everything else in Compartment.
• CLI for everything else - scripts, cron, other agents: compartment store, compartment search, compartment recent, compartment forget, compartment link, compartment relations, compartment import-claude, compartment lock.
• See the vault: compartment dash - one command opens a local page with everything at a glance: how many memories of what kind, growth over time, the relation graph, tags, per-agent counts, live search. Served from RAM, 127.0.0.1-only behind a random URL token, read-only, zero outbound requests, zero configuration.
• Panic lock from the agent - memory_lock / compartment lock clears stored credentials instantly when you need the vault closed now.
• One vault, many hosts - Hermes, Claude, and the CLI can share a vault at once; each caller gets its own identity and namespace ACLs.
• One memory, no sections - the starting memories seeded at init live in main as ordinary records, editable and forgettable like anything the agent stores; older vaults reorganize automatically.

Day to day, the point is simple: the agent remembers you, your decisions, and your machine - encrypted, offline, and fast - without a cloud account.

The lock model

You lock and unlock the vault yourself whenever you want. Manual control is always available:

• compartment unlock - open the vault with YOUR passphrase. You choose it; Compartment never auto-generates a password, seed, or recovery phrase, and there is no credential it knows that you don't. (Vaults made by older versions that received an auto-generated recovery phrase still open with it.)
• compartment lock - close it again and clear every stored credential. Agents can do the same via the memory_lock panic tool.
• compartment 2fa enable - optional two-factor unlock: your passphrase (knowledge) plus a keyfile (possession - keep it on a USB stick). Both factors feed Argon2id together, so needing both is enforced by arithmetic, not a policy check; a stolen vault file plus your passphrase still opens nothing without the keyfile. One command, zero configuration: the keyfile's location is remembered, so day-to-day unlocking feels exactly the same while the file is present.

The default unlock mode is convenience, not a cage: after a normal unlock, the vault stays usable across processes, logouts, and logins - for weeks or months if you leave it that way - until the next restart or power loss, or until you lock it yourself. Restart/power loss always locks it: the stored credential is the master key wrapped under a random 32-byte per-boot secret, held in a volatile kernel object that is never written to any filesystem, so a restart destroys it and a new boot can never open the old wrap. A copy of the credential file on its own is useless, because the key it needs was never on the disk. That is arithmetic, not a policy check.

If you prefer reboot-surviving unlock on macOS, that is an explicit opt-in (compartment unlock --keychain), with the tradeoff documented. At any time you can lock, unlock, lock again - on your schedule.

Security, in one paragraph

XChaCha20-Poly1305 AEAD on everything at rest including vectors (embedding-inversion resistance) · Argon2id keyslots, LUKS-style, opened only by the user's own passphrase (no auto-generated credentials), optionally two-factor with a keyfile · per-record keys enabling forget --shred (crypto-shred: key destroyed, content mathematically unrecoverable) · fsync'd sealed journal, atomic compaction, verified kill-9 crash recovery · hash-chained tamper-evident audit log (compartment audit verify) · per-caller namespace ACLs, quarantine tier for untrusted content, signed vault manifests · stdio MCP transport: zero open ports · runtime offline guard that aborts on any socket attempt; CI runs the whole suite with it active on Linux, macOS, and Windows · no telemetry, ever. Full honest threat model, including what Compartment cannot protect against, in SECURITY.md.

One vault, many agents

Hermes, Claude, and the CLI can share a single vault simultaneously: writes are serialized by an advisory file lock, every process detects foreign writes and reloads, and each host gets its own caller identity and namespace with rw/ro grants. A locked vault is one portable file, safe to move over any channel; compartment lock --sign seals it with an Ed25519 manifest the recipient can verify without any credential.

compartment lock
scp ~/.compartment/memory.vault other-machine:
compartment --vault memory.vault unlock     # your passphrase (+ keyfile if 2FA)

Configuration

Nothing here is required. Compartment installs configured, and this is the whole surface if you want to change something.

In the app

The panel behind the icon: Unlock and Lock, Change password, Create memories automatically (the capture hook), Search starter facts, Auto-lock (15, 30, 60 minutes or never), the CONNECT AN AGENT buttons for Claude, Hermes and OpenClaw, Refresh and Quit.

compartment panel --login on | off | status controls starting at login, which on Linux is the applications menu entry.

Commands

Command	What it does
init	create the vault. --passphrase, --creator, --keychain, --no-session, --no-app
unlock / lock	open or close it. --passphrase-stdin, --keyfile, --keychain, --once; lock --sign --identity
status / verify / selftest	what is in it, is it intact, does it work
store / get / forget	one memory. --namespace, --tag, --importance, --quarantined, forget --shred
search / recent	find things. --namespace, --tag, --top-k, --limit, --all, --json
link / relations / unlink	the relation graph, with validity windows (--from, --to, --as-of)
panel (menubar, tray)	the app. --show, --self-check, --render, --login
integrate <agent>	wire claude, hermes or openclaw, and install /compartmentalize for it. --no-import, --no-hooks
hook	capture hook: install --pin-vault, uninstall, status, capture
serve	the MCP server, over stdio
dash	read the vault in a browser: 127.0.0.1, one-time token, GET only
export / import	export --plaintext writes it unencrypted; import reads it back
import-claude	pull in what Claude Code already wrote. --dir, --namespace, --dry-run
rekey	change the passphrase. --new-passphrase-stdin
2fa	enable, disable, status - a keyfile as a second factor
audit	verify, repair the hash-chained history
pack	build, install, remove, list, export signed memory packs (--trusted-key)
reindex	rebuild the index, and give long records the embedding windows they are missing. --int8, --f32, --re-embed, --model
bench	--records, --longmemeval, --variant, --limit
setup	download-model, download-longmemeval, airgap-bundle
update	upgrade in place. --source takes GitHub main, --no-app skips the restart
uninstall	remove it. The vault is kept unless you pass --purge

Global flags, before the command: --vault PATH, --caller NAME, --keyfile PATH, --assert-offline, --version.

The /compartmentalize skill

compartment integrate <agent> writes one file into that agent's own skills directory, and compartment uninstall takes it back:

Agent	Path
Claude Code	~/.claude/skills/compartmentalize/SKILL.md
Hermes	$HERMES_HOME or ~/.hermes/skills/compartmentalize/SKILL.md
OpenClaw	$OPENCLAW_HOME or ~/.openclaw/skills/compartmentalize/SKILL.md

All three read the same Agent Skills layout, so it is one packaged file. It is user-invoked only: no agent runs it on its own guess. Edit your copy freely - a later install backs up anything that differs rather than overwriting it, and leaves the backup behind when the skill is removed. Invoking it makes the agent sweep the conversation and write to the vault, so expect a burst of memory_store calls; that is the point of it.

Settings file

<vault>.config.json, beside the vault, holding grants per caller and:

Setting	Default	Meaning
auto_lock_minutes	30	idle time before it locks. 0 never locks
search_starter_facts	true	whether the seeded facts join search results
include_packs_in_search	true	the same, for installed packs
duplicate_threshold	0.97	cosine similarity at which a store is a duplicate
index_precision	"f32"	"int8" uses a quarter of the RAM
unlock_tool_enabled	false	lets an agent unlock the vault. Off because the passphrase would cross the model's context

Environment

COMPARTMENT_VAULT which vault to use, COMPARTMENT_PASSPHRASE for scripts and CI, COMPARTMENT_SESSION_DIR where the unlock credential lives, COMPARTMENT_UI_SCALE panel scale, COMPARTMENT_ASSERT_OFFLINE abort on any network attempt. HERMES_HOME, OPENCLAW_HOME and XDG_DATA_HOME are read where they apply. Anything exported as ENGRAM_* still works.

MCP tools

memory_search, memory_store, memory_get, memory_recent, memory_forget, memory_link, memory_relations, memory_unlink, memory_list_namespaces, memory_status, memory_lock, memory_selftest. memory_unlock exists but is off unless you turn it on above.

Documentation

docs/MEMORY.md	how memory is stored, what gets remembered, why the math wins
docs/INTEGRATIONS.md	selecting Compartment in Hermes, OpenClaw, Claude, everything else
SECURITY.md	full threat model, honest limits
FORMAT.md	byte-level .vault and .mpack specs (language-agnostic)
PACKS.md	authoring and shipping signed memory packs
RELEASING.md	cutting a release: every download, every time

---

mcp-name: io.github.MaxFreedomPollard/compartment