Pseudolife Memory

Pseudolife-MCP

简体中文 · 日本語 · 한국어 · Português (BR) · Español

Persistent long-term memory for Claude Code, Codex, and other MCP clients.

An MCP server that gives coding agents a long-term memory that persists across sessions — surviving context compactions and fresh tasks. Your coding agent is the intelligence; this server is its memory on disk.

What you get:

• Associative memory with honest forgetting — a flat similarity store ranked by hybrid dense-plus-lexical retrieval, with conflict detection that admits potential updates while preserving earlier source notes; whole-note replacement is explicit. (The measured verdict: a preregistered ablation campaign found the previous 8-band continuum tied a flat store on every gate, so the simpler structure ships; the continuum remains one config line away.)
• Canonical facts, not vibes — one current value per entity.attribute slot (or a member set, for slots that hold many concurrent values); corrections supersede rather than silently overwrite, and the full version history survives.
• Dreams — a bundled local extractor, or any OpenAI-compatible endpoint (a Claude model on your Max plan, a GPT-5.6 model on a ChatGPT plan, LM Studio, Ollama, vLLM), consolidates the memory stream into facts and a knowledge graph while you're not looking.
• Lessons from its own work — successes, dead-ends, and your corrections become do/avoid guidance surfaced at the start of every session.
• A web console to watch it think — the Cortex Console above, plus cited world facts, session episodes, and document RAG.

Measured, with receipts — the full 500-question LongMemEval sweep, all six question types, and every number ships with its committed run artifact:

Equal accuracy to naive RAG across the whole benchmark on ~73% of the context, and better calibrated about what it does not know: on BEAM-100K's abstention questions the fact spine scores 0.950 against naive RAG's 0.775, unchanged under two independent judges. Read that as calibration, not recall — in the budget-matched five-arm run of 2026-09-02 (rag 0.725 there; one replicate, local judge) an arm served no memory at all scores 1.000 on the same questions, because refusing is the right answer there and an empty context always refuses. The fact spine loses where an answer has to be aggregated across sessions. The second claim to survive a judge swap is a win rather than a wash: re-run on 2026-09-04 with the hybrid arm budget-matched to the control at 6 turns, the same 500 questions give hybrid 0.730 against naive RAG's 0.690 under the local judge and 0.736 against 0.694 under claude-opus-5 — paired +0.040 / +0.042, p 0.015 / 0.013 — bought with more context, ~1229 tokens against the control's ~1124, not less, and carried mostly by temporal-reasoning questions. Graded by a local, byte-reproducible judge (the cross-judge check names its second judge) — compare within rows, never against GPT-judged leaderboards.

Retired 2026-08-25 (#188): the 0.936 knowledge-update headline. It was measured on the 2026-07-30 bench stack (Qwen3.6-27B answerer and judge). Re-running the same 78 questions after the 2026-08-17 migration to Qwen3.8-27B puts the cascade at 0.846, below the naive-RAG control — which lands on 0.859 on both stacks. The cascade serves the fact-spine answer unless that channel says "I don't know", so it measures the answerer's abstention behaviour as much as the memory: 32/78 abstentions at 46/46 commit precision on the old stack, 22/78 at 0.839 on the new one. The 500-question table above is on the older judge and has not been re-judged, so read its cascade row as an upper bound.

Full tables, the per-type breakdown, both stacks side by side, and every artifact: Benchmarks.

Quickstart

Install and register the lite tier. No Docker, no database to set up, no container runtime:

pip install "pseudolife-mcp[lite]"
claude mcp add --scope user pseudolife-memory -- pseudolife-mcp

Codex instead of Claude Code — same shape:

pip install "pseudolife-mcp[lite]"
codex mcp add pseudolife-memory --env PSEUDOLIFE_WRITER_ID=codex -- pseudolife-mcp

For Codex, finish setup before starting a fresh task. In the existing [mcp_servers.pseudolife-memory] table in ~/.codex/config.toml, add startup_timeout_sec = 240, tool_timeout_sec = 240, and required = true. The shim can wait up to 180 seconds for a cold daemon; Codex's default startup budget is 10 seconds. required makes missing memory visible at startup and waits for its initial catalog. These are starting budgets, not a promise that a first model download fits. The tool budget leaves time for the shim's 180-second deadline to report a failure before the host cancels it; prewarm with pseudolife-mcp serve in a terminal if needed.

Approve memory_message in that table's tool configuration ([mcp_servers.pseudolife-memory.tools.memory_message] approval_mode = "approve", or allow it once and keep the approval); Codex hook setup (python ops/setup-codex-hooks.py) sets it when you approve the hooks, and keeps any value you chose. Board mail can wake an idle Codex task by default, and the woken task reads its mail with that tool: without the approval it stalls on a prompt until someone answers it. Set PSEUDOLIFE_CODEX_DOORBELL = "0" in the same env table to keep the task from being woken (see Codex doorbell).

The agent board (peer awareness and addressed mail between sessions) is on by default, behind bearer authentication. Without coordination.allowed_principals in the daemon's config.yaml, only the singular PSEUDOLIFE_MCP_TOKEN principal is admitted. A Codex bearer from a PSEUDOLIFE_MCP_TOKENS map stays off the board until an operator lists it (allowed_principals: [default, codex]). Until then a shim on the default setting leaves coordination off without an error; one pinned with --enable shows an attach-unavailable hint instead. python ops/setup-codex-coordination.py --check reports ready (default-on) or names the cause. See Codex CLI and desktop.

The MCP handshake delivers compact recall/capture/reflection instructions. For the complete standing guidance, copy the bundled memory block into your project AGENTS.md or ~/.codex/AGENTS.md. For session briefings and per-turn reminders, follow Codex hooks and verification. Use one MCP registration and one hook source; an installed plugin may already provide either. After the daemon is running, execute pseudolife-mcp doctor from the same environment as the registered command. It checks the handshake and annotations without calling bank tools. When the shell has no PSEUDOLIFE_MCP_TOKEN or PSEUDOLIFE_MCP_TOKEN_FILE, doctor takes the token from the Claude Code registration (~/.claude.json, or under CLAUDE_CONFIG_DIR), else from the Codex one, and names where it came from in credential_source; with none and a daemon that requires a token, it reports BearerMissing.

Then in either coding agent: "remember that my staging box is haze-02" → the agent calls memory_store; next session, "which box is staging?" → memory_search finds it. Browse everything at the Cortex Console: http://127.0.0.1:8765/ui/.

The first session auto-starts the daemon, which provisions an embedded PostgreSQL 18 (pgvector included, via pg0-embedded) under a stable per-user data dir and downloads the embedding model (~1.2 GB, one-time). It is a real Postgres bank, not a cut-down one: pseudolife-mcp backup writes a standard owner-free pg_dump archive (plus a state archive, 7-day rotation) that restores into any PostgreSQL 18 target regardless of role — the Docker tier included — so outgrowing lite is a dump/restore, not a migration project (backups). For a tier- and Postgres-version-independent copy, pseudolife-mcp export / import move the whole bank as portable JSONL (logical export / import). Windows needs an ASCII-only data path (PSEUDOLIFE_MCP_DATA_DIR).

What lite gives you, and the one thing it doesn't

The gap, stated plainly. Lite ships no extractor, so the dream pass still runs, prunes, and acknowledges its input batch, but writes no canonical facts: on this path memory_fact_set is the only cortex writer. Everything else above works. Nothing about this is silent — curl http://127.0.0.1:8765/health reports "extractor": "none", and the stdio shim says the same on stderr at session start.

Any OpenAI-compatible endpoint closes it. The daemon inherits the environment it starts from, so two variables are the whole fix — with a local Ollama:

export PSEUDOLIFE_DREAM_BASE_URL=http://localhost:11434/v1
export PSEUDOLIFE_DREAM_MODEL=qwen2.5:7b
pseudolife-mcp serve

$env:PSEUDOLIFE_DREAM_BASE_URL = "http://localhost:11434/v1"
$env:PSEUDOLIFE_DREAM_MODEL    = "qwen2.5:7b"
pseudolife-mcp serve

/health then reports "extractor": "configured". One gotcha: a daemon that is already running keeps the environment it started with, and the shim reattaches to it rather than spawning a new one — stop the old daemon first. A hosted endpoint works too, and costs you the zero-egress property: memory text leaves the machine. Extractor tiers, quality, and the trade-offs: Dreaming.

Durable tier — Docker (recommended for a long-lived bank)

Everything above plus the bundled extractor, external volumes, health-checked services, and backup/rollback tooling. Requires Docker and at least one MCP-capable coding agent — Claude Code, Codex, and Gemini CLI are wired end-to-end; anything else gets paste-ready config (provider matrix). One command from clone to first memory:

git clone https://github.com/Pseudogiant-xr/Pseudolife-MCP.git
cd Pseudolife-MCP
ops/install.sh          # Linux / macOS
ops\install.ps1         # Windows (pwsh 7+)
# Codex: add --client codex / -Client codex
# Codex defaults to automatic hook-source detection and asks once for approval.
# Unattended hook approval: --codex-hook-trust yes / -CodexHookTrust yes
# Instructions only: --codex-hooks skip --instructions append
# PowerShell equivalent: -CodexHooks skip -Instructions append
# Both:  add --client both  / -Client both
# Gemini: add --client gemini — or several: --client claude,codex,gemini
# Other MCP agents (Cursor, Windsurf, Zed, ...): --client generic

The installer first asks where the memory bank lives: on this machine (the default, and everything below), on another machine that already runs it (a client-only install that asks for the daemon's URL and token), or on this machine with other machines connecting to it (see Sharing one bank across machines). It then asks which agents to wire (multi-select, with a capability matrix showing exactly what each one gets — session briefing, per-turn discipline, standing file), runs the preflight (one exact fix line per missing prerequisite), then asks which dream extractor should consolidate memories —

• sidecar — the bundled local CPU model; no Claude plan needed, works for everyone, and keeps every memory on the box (~11.8 GB image);
• claude-only — the lightest install: a Claude model via a CLI shim (claude-opus-5-5 by default since 2026-09-29, when it cleared the extraction-ladder gate against claude-opus-5; needs a logged-in Max-plan claude CLI); the sidecar image is never built or pulled (~11.8 GB lighter; dreams pause while the shim is down);
• claude-fallback — the Claude shim primary, the bundled sidecar as automatic fallback (Max-plan CLI plus the ~11.8 GB image);
• openai-only / openai-fallback — the same two shapes on an OpenAI subscription: a GPT-5.6 or GPT-6 model (Terra by default) via the Codex CLI shim on a signed-in ChatGPT plan (extraction quality unmeasured — see the dreaming guide);
• endpoint / endpoint-fallback — any OpenAI-compatible server you name (--extractor-url, --model: LM Studio, Ollama, vLLM, a hosted API), alone or with the sidecar as fallback. The earlier names sonnet-* and codex-* still work as deprecated spellings —

then brings the stack up, installs the selected clients' session hooks (where the client has a hook system), registers the MCP transport (the stdio shim by default, with a per-provider writer id; direct HTTP via --transport http), and health-checks the daemon — finishing with a per-agent ladder of what got wired and what that agent's platform cannot support. Codex setup offers one choice to enable automatic memory briefings, reminders, and session cleanup (plus, where the agent board is on, a board check-in at session start and a new-mail hint per prompt), use standing instructions only, or skip. Automatic setup reuses an enabled PseudoLife plugin or installs the three lifecycle hooks, backs up configuration, approves only their exact current definitions, and verifies execution. If verification fails, the same approval allows the standing memory block as a fallback; setup reports the remaining repair step. Hook-less providers (Gemini CLI and generic agents) are offered the standing block. --instructions append always writes the block from examples/CLAUDE.memory.md into ~/.claude/CLAUDE.md / ~/.codex/AGENTS.md / ~/.gemini/GEMINI.md (useful for subagent visibility even with hooks). Idempotent — re-run any time; --extractor <mode> switches extractor setups. Non-interactive example: ops/install.sh --extractor sidecar --client codex --codex-hook-trust yes. Without explicit hook approval, unattended setup does not grant trust; use --codex-hooks skip --instructions append for instructions only. Explicit --instructions skip prevents fallback edits. The installer also turns the agent board on: it mints a bearer token in ops/.env (owner-only, never printed) and gives each shim client it wires an owner-only token file. Re-running it on an existing install adds that file to the Claude Code registration in place. The ladder's last line, and pseudolife-mcp doctor, say whether the board is on or why it is off. Use doctor --host codex or --host claude-code for a compact coordination snapshot: reachability, bearer admission, registration evidence, tool inventory and configured wake are separate. Default doctor is read-only; health and a configured wake path do not prove delivery. The coordination diagnostic quickstart covers saved-instance checks and an explicitly disposable mail proof. --no-token / -NoToken keeps an open-loopback install with the board dormant (Turning the board on). Linux (Docker Engine): your user must be in the group — , then log out/in (the preflight checks this).

Image sizes, the Windows WSL2 memory cap, and what the installer automates: the containerized install below.

ops/preflight.sh --client codex    # or ops\preflight.ps1 -Client codex
docker volume create pseudolife-mcp-bank
docker volume create pseudolife-mcp-state
docker compose -f ops/docker-compose.yml up -d --build   # first build, once

# ...or pull the prebuilt images instead of building (releases >= 0.14.0):
docker compose -f ops/docker-compose.yml -f ops/docker-compose.ghcr.yml pull pseudolife-pg pseudolife-daemon
docker compose -f ops/docker-compose.yml -f ops/docker-compose.ghcr.yml up -d

# Verify, then wire the transport into one or both clients.
curl http://127.0.0.1:8765/health

# Stdio shim (the installer's default — per-session episode identity).
# PSEUDOLIFE_MCP_NO_SPAWN=1 makes the shim wait for the container instead
# of spawning a host fallback that can shadow the Docker bank after a
# reboot; set it on Docker-tier registrations like these.
pip install pseudolife-mcp
claude mcp add --scope user pseudolife-memory --env PSEUDOLIFE_MCP_NO_SPAWN=1 -- pseudolife-mcp
codex mcp add pseudolife-memory --env PSEUDOLIFE_MCP_NO_SPAWN=1 -- pseudolife-mcp

# ...or direct HTTP (no pip package needed; fine for single-session setups):
claude mcp add --transport http --scope user pseudolife-memory http://127.0.0.1:8765/mcp
codex mcp add pseudolife-memory --url http://127.0.0.1:8765/mcp

# Reinforce the protocol-level memory loop with a global standing instruction:
cat examples/CLAUDE.memory.md >> ~/.claude/CLAUDE.md
cat examples/CLAUDE.memory.md >> ~/.codex/AGENTS.md
# (PowerShell: Add-Content "$env:USERPROFILE\.claude\CLAUDE.md" (Get-Content examples\CLAUDE.memory.md -Raw))

Optional knobs live in ops/.env (cp ops/.env.example ops/.env — the install/update scripts scaffold it too; every value is commented, a missing file runs entirely on defaults).

What this is

A memory engine exposed over MCP. There's no chat UI and no LLM doing the thinking — your coding agent is the intelligence; these are tools it calls to store and recall what matters. (Models are bundled as plumbing: baked embedding weights for retrieval, and the optional CPU extractor sidecar that consolidates memories into facts while you sleep.)

Where it sits among the common approaches to agent memory — each column is a fair tool for what it's for; this table is about what question each one answers, not who's wrong:

Auto-journaling records what the agent did; Pseudolife curates what it learned. Both are useful — they answer different questions. Named alternatives — Mem0, Zep/Graphiti, Letta, Cognee, memU, Memori — and the cases where one of them is the better pick: Comparison.

It layers several complementary stores: the associative store (a flat embedding store ranked by cosine similarity fused with a BM25 lexical pool (on by default), with conflict-aware admission and explicit source-note replacement; an 8-tier banded layout is available as an opt-in preset); the cortex (slot-keyed canonical facts — one current value per entity.attribute, or a member set for set-valued slots — with provenance tiers and contender parking instead of silent overwrites); a typed knowledge graph over those facts with a closed relation vocabulary and on-read inference; the world cortex (durable cited facts about external reality, age-decayed trust); procedural lessons learned from the agent's own work; and a ChromaDB reference bank for document RAG. The canonical layers in depth: the memory model; the graph and multi-hop recall: retrieval.

State lives in Postgres (the durable source of truth) behind a single long-lived daemon; every session attaches through a thin stdio shim (installer default — per-session identity) or directly over HTTP (single-session setups). The result: Claude can pick up where it left off, correct itself when facts change, and reason over relationships — without you re-explaining context each session.

Documentation

This README is the front door — install, wiring, and the basic loop. The deep material lives in the user guide:

Plus evals/README.md (full benchmark methodology) and CONTRIBUTING.

Tools exposed

The surface was consolidated 2026-07-02 (55 → 32 tools; now 38 with memory_toolset, the set-slot pair and coordination): lifecycle families became verb-dispatched tools (memory_dream, memory_forget, memory_graph_review), and dump/introspection views moved to the Cortex Console (REST) — the manifest is agent context every session, so it stays lean.

Each tool returns plain JSON. See pseudolife_memory/mcp_server.py for docstrings — those are what Claude reads to decide when to call which tool. The five recall-path tools return compact entries by default (result payloads are agent context on every retrieval; search and recent entries keep their write date), and memory_fact_get likewise drops bookkeeping keys from its record; pass verbose=true for full metadata. Full-table dumps and topology views live in the Cortex Console (/api/*) and the pseudolife-mcp briefing CLI.

Toolset tiers. Three visibility tiers — minimal (10 tools), core (24), full (38) — filtered per principal at tools/list; a principal (the named bearer-token identity, or the writer id for single-token installs) steps its own tier up or down with memory_toolset before calling a hidden tool. Defaults, per-client mapping, and weak-model deployments: Configuration — toolset tiers.

Architecture

One memory daemon owns the bank and serves MCP over streamable HTTP at /mcp; every Claude Code session (and any LAN agent) attaches to it. Postgres 18 + pgvector (in Docker on the durable tier; the lite tier runs the same Postgres embedded, no container) is the durable source of truth — the in-memory store is a write-through cache hydrated at startup (a small weights.pt persists only counters — there are no MLP weights).

The daemon runs either containerized (recommended — portable, no host Python) or as a host process. Claude Code attaches through a thin torch-free stdio shim (the installer default — per-session identity, needed for concurrent sessions) or directly over HTTP (simpler for a single session):

Claude session A ─┐  stdio shim (installer default) or HTTP
Claude session B ─┼───────────────────► pseudolife-mcp daemon ─► Postgres (Docker)
LAN agent ────────┘  or stdio shim         (single writer)        pgvector
                     (per session)         host proc OR Docker

This kills two v0.1 hazards by construction: a single writer means concurrent sessions can't clobber each other, and entries are transactional so a crash can't wipe the bank. The single writer is enforced: the daemon holds a Postgres advisory-lock writer lease on its bank. A second daemon, or a script or eval that opens the live bank through the service, refuses to start and names the process that holds it. Stop the daemon before offline maintenance such as ops/dedup_cortex.py. If the bank fails to load at startup, the daemon serves nothing rather than a partly loaded bank, and /health reports degraded with the reason until a retry succeeds. On top of the associative store sit the canonical layers — cortex, world facts, lessons, temporal/HLC stamps (the memory model) — joined to a typed knowledge graph walkable via memory_graph and multi-hop memory_recall (retrieval & the graph).

Install — containerized (any OS)

What the durable tier installer above does, by hand. The whole stack — Postgres and the memory daemon — runs in Docker. No host Python, no torch install, no version skew; the daemon image bakes in CPU-only torch and the embedding weights — Qwen/Qwen3-Embedding-0.6B (the default retrieval backbone since schema v25) plus all-MiniLM-L6-v2 (kept baked for the ONNX-parity test path) — so it runs identically on Windows / macOS / Linux. Requires only Docker; built once: ~5.0 GB daemon image (measured 2026-07-29 on the deployed build) + ~0.6 GB Postgres + ~11.8 GB extractor sidecar (measured 2026-08-20 with the v3 multi-task bake; skip the sidecar entirely with the installer's claude-only mode). The ~12.6 GB and ~10.4 GB figures published before 2026-07-29 are retired: both were inflated by a CUDA torch build that a dependency-resolution bug pulled into the image (see the CHANGELOG); the daemon has always been CPU-only.

git clone https://github.com/Pseudogiant-xr/Pseudolife-MCP.git
cd Pseudolife-MCP

# 1. One-time: create the two persistent volumes (bank + daemon state).
docker volume create pseudolife-mcp-bank
docker volume create pseudolife-mcp-state

# 2. Build + start all three services (Postgres, extractor, then the daemon).
docker compose -f ops/docker-compose.yml up -d --build

Or skip the ~5 GB daemon build entirely and pull the prebuilt images (releases ≥ 0.14.0):

docker compose -f ops/docker-compose.yml -f ops/docker-compose.ghcr.yml pull pseudolife-pg pseudolife-daemon
docker compose -f ops/docker-compose.yml -f ops/docker-compose.ghcr.yml up -d

The extractor sidecar is not published and still builds locally; update the pull path with pseudolife-mcp update (backup, pinned image, rollback tag; see Updating), not a bare pull + up -d.

Upgrading from a pre-rename install (volumes ops_pseudolife_pgdata / ops_pseudolife_data)? Don't rename those volumes — keep pointing at them by creating ops/.env with PSEUDOLIFE_BANK_VOLUME=ops_pseudolife_pgdata and PSEUDOLIFE_STATE_VOLUME=ops_pseudolife_data before up. See the compose header.

Windows: cap Docker Desktop's WSL2 VM, which otherwise claims up to ~50% of host RAM — how much the stack actually needs, the ops/wslconfig.example template, and the daemon container's own memory cap: Configuration — Windows / WSL2 memory.

The daemon serves MCP at http://127.0.0.1:8765/mcp and restarts with Docker — no logon task needed. First build downloads the model into the image (once); every container start after that is offline and fast. Wire Claude Code in via the stdio shim (installer default) or directly over HTTP (both below). Where the data actually lives, and how to back it up: Configuration — data layout.

Other machines, one bank: sessions on a laptop, a server or a container can use this daemon too, through a client-only install that runs just the shim. Exposing the daemon (Tailscale Serve, a LAN publish or a reverse proxy), per-machine tokens and board admission: Sharing one bank across machines.

Host-process install (Windows, for GPU / dev): run Postgres in Docker but the daemon on host Python — for hacking on the daemon or running the embedder on a local GPU. Steps, the pseudolife-mcp CLI modes, and the logon autostart task: Configuration — host-process install.

Updating

Upgrading from before the agent board (no bearer token yet): this migration step applies to installer-managed Docker installations. Rerun ops/install.ps1 (Windows) or ops/install.sh (Linux / macOS) with the same client selection (sessions can stay open: the shim installs as a new runtime beside the running one). The installer creates a bearer token for the default shim install and migrates the environment of the installer-managed pseudolife-memory stdio registration for Claude Code in place. Custom registrations are preserved; follow the installer's printed credential warnings for Gemini or custom registrations. -All / --all and ops/update_clients.py do not create the token or migrate the registration environment. Then refresh clients with the Everything at once recipe below and restart them.

A plugin marketplace added as Pseudogiant-xr/Pseudolife-MCP (installers up to 0.16.1) is refreshed over SSH, which fails on a machine with no GitHub SSH key. pseudolife-mcp update and the installers move it to the HTTPS URL themselves, with backups. If that step reports a failure, the manual fallback is: in settings.json ($CLAUDE_CONFIG_DIR, else ~/.claude), set the pseudolife-mcp entry's source under extraKnownMarketplaces to {"source": "git", "url": "https://github.com/Pseudogiant-xr/Pseudolife-MCP.git"}, then run claude plugin marketplace add https://github.com/Pseudogiant-xr/Pseudolife-MCP.git (never marketplace remove it: that uninstalls the plugin).

One command, no checkout: the installed shim updates the whole install from a release:

pseudolife-mcp update            # the newest release on PyPI
pseudolife-mcp update --tag 0.15.1
pseudolife-mcp update --check    # exit 0 when a newer release exists, 3 when current

The first update from 0.15.0 or earlier has no such command (that shim answers "unknown mode", and that daemon never announces a release): from a checkout, git pull, then ops/update.sh --all or ops/update.ps1 -All; for a pipx or pip install, pipx install --force "pseudolife-mcp[lite]==<version>" or pip install --upgrade "pseudolife-mcp[lite]==<version>" (the [lite] extra only for a lite install). After that, pseudolife-mcp update exists. On a client-only machine (the daemon on another host), update and --clients-only move this machine's clients to the daemon's release.

On a Docker-tier install it pulls the pinned GHCR daemon image, backs the bank up (the checkout's backup script when the compose project still has one, else its own pg_dump + state-volume tar into ~/.pseudolife-mcp/ backups), tags the running image for rollback, recreates only the daemon container from the compose files it was created with (or the package's bundled copies when the checkout is gone), waits for /health at the new version, then installs the release as a new shim runtime beside the running one, refreshes the plugin cache and prints the Codex hook step. The backup stays visible and deliberate: nothing here automates it away. --clients-only / --daemon-only take one half; the shim itself runs from wherever it was registered, so use its full path when pseudolife-mcp is not on PATH. On a pip / lite install it upgrades the package in the interpreter that holds it (pip, or pipx) and restarts nothing: it says what to restart.

You are told when it is time: the daemon checks PyPI for the newest release in the background (updates.check_releases, on by default) and every session's briefing then opens with release X is available — run pseudolife-mcp update. With updates.unattended_clients: true in config.yaml (off by default) a Docker-tier shim on the daemon's host installs the daemon's release as its own new runtime and refreshes the plugin cache by itself whenever the daemon is newer than it; the daemon recreate stays a deliberate command. On a headless host, pseudolife-mcp update --schedule 03:30 installs a daily task or timer that applies a new release only while updates.unattended_daemon: true is set and no session is active on the agent board, with the same backup and rollback tag, and posts a board notice either way. Codex keeps its hook approvals across updates (it approves the hook definitions, not the scripts), and an update from a checkout refreshes manual Codex hook copies itself. Where Codex runs the plugin's hooks, the client step refreshes Codex's installed plugin copy through codex plugin marketplace upgrade while the approved hook definitions are unchanged (Codex specifics); only a changed hooks.json makes every update path print the approval steps. See Updating.

Lite tier by hand: one command, bank untouched:

pip install -U "pseudolife-mcp[lite]"

On Windows, first close every Claude Code, Codex and Claude Desktop session using the shim (quit Desktop from the tray): upgrading a package that is running can leave it half-removed.

Docker tier from a checkout: after a git pull (or local code change), redeploy the daemon only — safely, without touching Postgres or the extractor:

.\ops\update.ps1        # Windows

./ops/update.sh         # Linux / macOS

Both are thin wrappers over the same Python deploy pseudolife-mcp update runs (ops/update.py → pseudolife_memory/update_cli.py), so there is one implementation. It backs up the bank (pg_dump + a state-volume tar), tags a rollback image (when a previous one exists — it says so loudly when there isn't), rebuilds + recreates only the daemon, and waits for /health. It never runs down -v. (Host-process install: just restart the daemon — pip install -e . is editable.) Build cache is pruned automatically after every healthy deploy; see Docker disk retention for the weekly Scheduled Task and the manual .vhdx compact. Never run docker system prune --volumes, which deletes volumes.

The image records the commit it was built from, and /health reports it as build (git_sha, dirty, built_at, and source: checkout or release). So the script refuses a tree with uncommitted or untracked files, and lists them. It also refuses a tree git cannot describe (no git, not a clone, or git's safe.directory refusal, which it quotes). Commit or clean up first, or pass -AllowDirty / --allow-dirty to deploy the tree as it is: stamped dirty: true, or unknown when git cannot describe it. Each deploy builds a new image, so the daemon container is recreated even when the commit has not changed.

Everything at once: the daemon is one of three installs. The shim your clients launch and the Claude Code plugin are separate and do not move with it. -All / --all moves them in the same run, after the daemon is healthy:

.\ops\update.ps1 -All   # Windows

./ops/update.sh --all   # Linux / macOS

No session has to be closed for the shim step. Each shim version installs into its own runtime (%LOCALAPPDATA%\pseudolife-mcp\runtimes\NNNNNN on Windows, ~/.local/share/pseudolife-mcp/runtimes/NNNNNN elsewhere) and every client registers one launcher path (%LOCALAPPDATA%\pseudolife-mcp\bin\pseudolife-mcp.exe / ~/.local/share/pseudolife-mcp/bin/pseudolife-mcp) that starts the newest complete runtime. The installer and the update step also make pseudolife-mcp typed in a terminal reach the launcher. On Linux and macOS ~/.local/bin/pseudolife-mcp becomes a link to it; an older pipx or pip --user entry there is moved aside as pseudolife-mcp.<kind>-<stamp> (never deleted), and anything else there is left alone and named. When ~/.local/bin is not on PATH, the step prints the line to add to your shell profile: export PATH="$HOME/.local/bin:$PATH". On Windows the launcher directory goes first on your user PATH, ahead of pipx's and pip's copies; open a new terminal for the name to reach the launcher, because a terminal that was already open keeps its old PATH. pseudolife-mcp doctor reports what the name resolves to (path_resolution) and warns when that is not the launcher; the launcher's full path always works. The step installs the checkout as a new runtime beside the old one, moves any registration that still names a runtime, pipx or virtualenv path to the launcher in place (Claude Code, Codex, Claude Desktop and Gemini CLI; each file is backed up first as <file>.bak-<stamp>), and removes older runtimes once no process runs from them and no registration names them. Sessions already running keep the runtime they started with; the next session start uses the new one. A shim running straight from this checkout's .venv is already live and is named instead. It then refreshes the plugin cache when its bytes differ from the marketplace clone (claude plugin update installs the new copy beside the one running sessions use; nothing is uninstalled), and reports whether Codex's hook copy matches the checkout (that refresh is a consent step: python ops/setup-codex-hooks.py). It ends with a ladder of what moved and which clients need a restart; a client-side step that fails is reported, never a failed deploy. The same helper runs on its own: python ops/update_clients.py, and python ops/shim_runtime.py manages the runtimes by hand (install, list, prune, migrate). Custom registrations are preserved and named; upgrade those in their own interpreter. A daemon-side credential change with an old shim leaves the two out of step: the Claude Desktop registrar refuses a shim that cannot read the selected token file (exit 4), and names the upgrade.

• Upgrading past 2026-09-28: board mail starts waking idle sessions. Once the plugin cache and the shim move (-All, then a client restart), a Claude Code session and a Codex task with a codex CLI are woken when mail arrives that clears the need they parked on; before this both wake paths were opt-in. Nothing rings for chatter, and rings are capped. To keep the old behaviour, set PSEUDOLIFE_AGENT_WAKE_HOOK=0 in the env block of ~/.claude/settings.json and PSEUDOLIFE_CODEX_DOORBELL = "0" in the Codex server's env table (PSEUDOLIFE_AGENT_COORDINATION=0 in either place turns off the board for that client). pseudolife-mcp doctor shows each client's wake path under wake.

The three tell on each other: /health reports the daemon's version and a digest of the hook scripts it shipped with; the session briefing opens with a one-line notice when the plugin's version differs from the daemon's (naming /plugin marketplace update pseudolife-mcp then /plugin update pseudolife-memory@pseudolife-mcp, or the daemon redeploy, whichever side is behind), or when the version matches but the hooks do not (naming the --all command above); the shim says so on stderr and ahead of its instructions; and pseudolife-mcp doctor reports version_mismatch.

Two upgrades are not automatic, because neither can be done safely in place. Both have a step-by-step runbook — backup, dry run, apply, verify, roll back — and a fresh install needs neither:

• A bank older than 0.11.0 (schema v25): every embedding column moved from vector(384) to vector(1024), so the daemon refuses to start rather than half-migrate. Re-embed offline with ops/migrate_embeddings.py — the v25 migration runbook.
• A Docker-tier bank created before 2026-08-14 (PostgreSQL 16 → 18): a Postgres major bump cannot reuse the old data volume. Run pwsh ops/migrate-pg18.ps1 — the PostgreSQL 18 migration runbook.

Wire into your coding agent

Plugin (hooks + commands). The installer adds it whenever Claude Code is a selected client (--claude-plugin skip / -ClaudePlugin skip opts out; an installed plugin is left alone). It wires the session hooks (briefing + episode identity), the memory-loop instructions, and the /dream + /memory-status commands. By hand, the same two commands inside Claude Code do it:

/plugin marketplace add https://github.com/Pseudogiant-xr/Pseudolife-MCP.git
/plugin install pseudolife-memory@pseudolife-mcp

The plugin replaces the settings.json hook, and the daemon serves a compact memory core and a live briefing as session context. By default (memory_policy.variant: compact) the full CLAUDE.md block below is not served; append it if you want the complete guidance. It deliberately does not bundle the MCP server: Claude Code loads a plugin server alongside any user-registered one with no deduplication, which doubled every session's tool namespace next to the installer's registration — so the transport is registered exactly once, by ops/install.* (stdio shim by default — per-session episode identity) or the one-liner below. Hooks an earlier install wrote to ~/.claude/settings.json would duplicate the plugin's; the installer offers to remove them once the plugin runs (--claude-legacy-hooks remove / -ClaudeLegacyHooks remove unattended). Details, non-default ports/tokens, and migration: plugin/README.md.

Manual transport registration. The installer's default (shim mode) registers a thin stdio shim — one shim process per session, so every session carries its own tier-1 identity. The same wiring by hand:

pip install pseudolife-mcp    # daemon in Docker; add [lite] for the pip tier
claude mcp add --scope user pseudolife-memory --env PSEUDOLIFE_MCP_NO_SPAWN=1 -- pseudolife-mcp

PSEUDOLIFE_MCP_NO_SPAWN=1 belongs on Docker-tier registrations: the shim then waits for the container instead of spawning a host-side fallback whose port bind can race a still-booting Docker and shadow the real bank. On the [lite] pip tier drop the --env — there the spawn fallback is the zero-config path.

Direct HTTP works too — the daemon serves MCP over HTTP natively (no shim, no host command, nothing OS-specific; concurrent sessions then share one episode identity, so it fits single-session setups best):

claude mcp add --transport http --scope user pseudolife-memory http://127.0.0.1:8765/mcp

(--scope user registers it for every project; drop it to register for the current project only.) Or write the equivalent JSON yourself — into ~/.claude.json under the top-level mcpServers key for user scope, or into a .mcp.json at a project root for project scope:

{
  "mcpServers": {
    "pseudolife-memory": {
      "type": "http",
      "url": "http://127.0.0.1:8765/mcp"
    }
  }
}

For a token-protected daemon, add a headers key to that Claude JSON entry: "headers": { "Authorization": "Bearer <your-token>" }.

Claude Desktop (the app, including its Cowork and Code sessions) — the installer writes the entry for you:

ops/install.sh --client claude-desktop      # Windows: ops\install.ps1 -Client claude-desktop

Desktop has no mcp add; its servers live in claude_desktop_config.json (macOS ~/Library/Application Support/Claude/, Linux ~/.config/Claude/, Windows %APPDATA%\Claude\ — except the Store/MSIX build, whose real file is under %LOCALAPPDATA%\Packages\Claude_*\LocalCache\Roaming\Claude\; the installer prefers that path when it exists). The equivalent entry by hand:

{
  "mcpServers": {
    "pseudolife-desktop": {
      "command": "/absolute/path/to/pseudolife-mcp",
      "env": {
        "PSEUDOLIFE_WRITER_ID": "claude-desktop",
        "PSEUDOLIFE_MCP_NO_SPAWN": "1",
        "PSEUDOLIFE_MCP_DAEMON_URL": "http://127.0.0.1:8765"
      }
    }
  }
}

The entry is named pseudolife-desktop on purpose. Desktop's Code tab runs Claude Code, which starts its own per-session pseudolife-memory server; when an app-level entry has the same name, Desktop sends the session's mcp__pseudolife-memory__* calls to the app-level entry and the session's own server gets none. Re-running the installer renames an entry it wrote under the old name (its env sets PSEUDOLIFE_WRITER_ID to claude-desktop), keeping any env keys you added and backing the config up first. It leaves any other pseudolife-memory entry alone and warns about it. Chat and Cowork then list the tools as mcp__pseudolife-desktop__*.

Two things differ from the CLI clients. Desktop launches MCP servers with a sanitized environment — PATH plus a few system variables, none of your shell's exports — so command must be the shim's absolute path (which pseudolife-mcp / Get-Command pseudolife-mcp), and a token-protected daemon needs "PSEUDOLIFE_MCP_TOKEN_FILE": "/absolute/path/to/a/private/file" in that env block: a bearer exported in your OS environment never reaches the shim. The symptom of forgetting it is every session failing with "Couldn't start for Cowork and Code sessions. Error: unhandled errors in a TaskGroup (1 sub-exception)" — a 401 under the SDK's wrapper, which the shim now names plainly on stderr at startup. When the daemon is token-gated the installer writes that file (owner-only) from PSEUDOLIFE_MCP_TOKEN in its environment or ops/.env, or migrates a literal token already in the entry into it; with no token to write it says so and exits 3. The shim must also be able to read that file: releases through 0.15.0 only read the literal PSEUDOLIFE_MCP_TOKEN, so the registrar probes <command> --help for the file form first and refuses an older shim (exit 4, nothing written) rather than register an entry that would fail with the same TaskGroup error — upgrade the shim (re-run the installer, or python ops/update_clients.py --only shim from the checkout: it installs a new runtime beside the running one, so no session has to close) and re-run. After any edit, fully quit Desktop from the tray or menu-bar icon and relaunch — closing the window does not reload the config.

Codex — the installer's default (shim mode) wires the same stdio shim, so a Codex session gets its own tier-1 identity instead of inheriting a concurrent Claude session's episode:

pip install pseudolife-mcp
codex mcp add pseudolife-memory --env PSEUDOLIFE_MCP_NO_SPAWN=1 -- pseudolife-mcp

(Same Docker-tier note as the Claude wiring above: keep PSEUDOLIFE_MCP_NO_SPAWN=1 when the daemon runs in Docker; drop it on the [lite] pip tier.)

The HTTP one-liner works too (no pip package needed):

codex mcp add pseudolife-memory --url http://127.0.0.1:8765/mcp

Or add the equivalent user-level entry to ~/.codex/config.toml:

[mcp_servers.pseudolife-memory]
url = "http://127.0.0.1:8765/mcp"
bearer_token_env_var = "PSEUDOLIFE_MCP_TOKEN"

For that Codex HTTP configuration, export PSEUDOLIFE_MCP_TOKEN in the environment that launches Codex. The token stays out of config.toml, and Codex reads it when connecting. This is unnecessary for the default stdio shim.

Gemini CLI — same shape (-s user matters: Gemini defaults to project scope; the -e env gives Gemini sessions their own write attribution, and the same Docker-tier PSEUDOLIFE_MCP_NO_SPAWN=1 note as above applies):

pip install pseudolife-mcp
gemini mcp add -s user -e PSEUDOLIFE_WRITER_ID=gemini -e PSEUDOLIFE_MCP_NO_SPAWN=1 pseudolife-memory pseudolife-mcp

Or HTTP, no pip package needed:

gemini mcp add -s user -t http pseudolife-memory http://127.0.0.1:8765/mcp

Note: since 2026-06-18 Google no longer serves individual-tier accounts (free, AI Pro, AI Ultra) through Gemini CLI — OAuth sign-in fails and points at Antigravity. The wiring above stays correct, but individual accounts need API-key auth (GEMINI_API_KEY) to actually run sessions — or use Google Antigravity itself, which connects to the same bank via ~/.gemini/config/mcp_config.json; both are covered in the providers guide.

Any other MCP-capable agent (Cursor, Windsurf, Zed, Copilot CLI, …) — add the generic mcpServers entry to that tool's MCP config (ops/install.sh --client generic prints both shapes ready to paste):

{
  "mcpServers": {
    "pseudolife-memory": {
      "command": "pseudolife-mcp",
      "env": {
        "PSEUDOLIFE_WRITER_ID": "mcp-client",
        "PSEUDOLIFE_MCP_NO_SPAWN": "1"
      }
    }
  }
}

What each agent gets — and what its platform can't support (hooks, per-turn discipline): the provider matrix.

Verify: run claude mcp list, codex mcp list, or gemini mcp list (the server should report connected), then ask the agent to "store a memory that this install works" and check it appears in the Console's Stream view at http://127.0.0.1:8765/ui/.

Preferring stdio (this is what the installer wires by default, for per-session identity)? A thin torch-free shim proxies stdio to the daemon: stdio shim · sharing one bank across machines · backups & restore rehearsal · agent mailbox recovery.

Recommended agent setup (CLAUDE.md / AGENTS.md)

The server's value depends on the agent using it. The MCP server advertises the core loop through protocol-level instructions; the shim adds the messageboard check-in when its coordination adapter is up. The plugin's memory SessionStart hook (also used by verified Codex hooks) delivers a short operating guide and a bounded briefing; without the plugin, the installer's Claude Code settings.json hook (pseudolife-mcp briefing --hook-json) delivers the same two. A separate coordination hook asks the agent to set its project, task and status, discover peers, and read pending messages, but only where the board is on for that credential (it is on by default behind bearer authentication, so an open install or a disabled board adds no check-in). Detailed memory guidance remains in the bundled standing block. Hooks add per-prompt reminders and session bookkeeping; neither delivery method guarantees that the model performs every requested memory operation.

By default (memory_policy.variant: compact) hooks serve at most that short guide, and the detailed block reaches an agent only as a standing copy (the full_separate_hook variant serves it in a separate hook; see Startup memory policy). For the complete guidance, for subagent visibility (subagents read CLAUDE.md but not hook output), or in place of hooks, append it to Claude's global ~/.claude/CLAUDE.md, Codex's global ~/.codex/AGENTS.md, Gemini's global ~/.gemini/GEMINI.md, or a per-project CLAUDE.md / AGENTS.md:

cat examples/CLAUDE.memory.md >> ~/.claude/CLAUDE.md
cat examples/CLAUDE.memory.md >> ~/.codex/AGENTS.md
cat examples/CLAUDE.memory.md >> ~/.gemini/GEMINI.md

Add-Content "$env:USERPROFILE\.claude\CLAUDE.md" (Get-Content examples\CLAUDE.memory.md -Raw)
Add-Content "$env:USERPROFILE\.codex\AGENTS.md" (Get-Content examples\CLAUDE.memory.md -Raw)

For hook-less providers this standing block supplies the full memory policy, but it cannot provide a live briefing or run session cleanup. AGENTS.md is the cross-vendor standard for standing agent instructions (Linux Foundation-governed; read by Codex, Copilot, Cursor, Gemini CLI, Zed, and 30+ others), so a per-project AGENTS.md carrying the block reaches almost every agent at once. Claude Code is the holdout — it reads CLAUDE.md — but a CLAUDE.md whose first line is @AGENTS.md imports the shared file, so one copy serves every tool.

The block (examples/CLAUDE.memory.md) teaches the loop: RECALL at the start (memory_search / memory_lesson_search / memory_fact_get / memory_world_search), CAPTURE as you go (memory_store with an honest origin, memory_fact_set for canonical facts, memory_world_set for cited external facts, source="status" for verbose logs so they stay out of the dream), REFLECT at the end (memory_outcome, with used_ids naming the hits you actually used — the dream distils these signals into the lessons surfaced at your next session start).

For an existing Codex installation, run python ops/setup-codex-hooks.py. The helper asks once, detects the hook source, backs up changed configuration, persists scoped trust through Codex, and verifies startup briefing, prompt reminder, and session cleanup, plus the agent-board check-in where the board is on. The Docker installer runs this step for you. See Codex setup options and fallback.

For Claude Code, use the plugin, or the legacy ops/install-hook.ps1 -Client claude / ops/install-hook.sh --client claude for briefing and reminder hooks. The legacy --client codex path remains available but only writes hook definitions; it does not complete trust and verification. Session episodes also work without hooks through the daemon: Episodes & sessions.

Current Codex runtimes enable hooks by default, including Windows. Availability depends on the application/runtime and policy, not the model. If [features] hooks = false is intentional, keep it and use the standing AGENTS.md block. Codex runs the plugin's Windows hooks as native PowerShell 7 commands; Claude Code runs their Bash commands through Git Bash, so it needs Git for Windows installed (see plugin/README.md). See the official hook protocol.

Codex hook trust: setup approval is limited to PseudoLife's current hook definitions: the memory and coordination SessionStart and UserPromptSubmit handlers and SessionEnd, plus, with the plugin, its Stop entry (Claude Code's wake hook, on by default; in Codex it runs only the park gate), its SubagentStart and two SubagentStop handlers (Claude Code's subagent board hooks, a no-op in Codex, plus the park gate) and its PreToolUse guard. It does not approve other plugins or bypass future trust checks. The same consent sets approval_mode = "approve" for the memory_message tool on the pseudolife-memory server, unless you set another value yourself, so a Codex thread can then read and send board mail without asking. Codex approves the definitions, not the scripts they run, so the approval also covers the scripts later PseudoLife updates install (for manual copies the update installs them itself, verified by content). Changed definitions need approval again, and so does a handler a plugin update adds; Codex skips an unapproved one silently in the desktop app, which pseudolife-mcp update (from a checkout, ops/update.ps1 -All or ops/update_clients.py) reports as needs-approval. If automatic setup cannot use the installed runtime's trust interface, it reports the problem and asks you to open /hooks to review and trust the definitions. Approved standing instructions remain available as fallback. Installed files alone do not establish that hooks are working.

Usage patterns

At session start — loads what you've worked on before, persistent across compactions:

memory_search("project context for X")

During work — store real decisions; skip fleeting chatter (the shipped store gate is permissive, so deliberate, durable claims only):

memory_store("Decided to use stdio transport for the MCP because no port conflicts", source="pseudolife")

When corrected — marks the old fact superseded and stores the correction; both surface in future retrieval, the new one ranked higher. Select the entry by the id carried on the search or recent hit:

memory_supersede(
  entry_id=417,
  new_text="Provider interface uses async calls — sync version was the v0.7 prototype only"
)

old_text= still selects by the full stored text when that text is exactly unique among live entries; it is the legacy selector and the only one file mode has. Ambiguous or missing targets change nothing.

Hygiene — memory and fact scopes hard-delete (at least one filter is required for scope memory, preventing accidental wholesale deletion; several filters narrow the match, so text="probe", source="status" is the probe entry in that source, not the whole source; a match over memory.delete_confirm_threshold entries — default 20, 0 disables — is refused with would_delete until the call repeats with confirm_bulk=true; on Postgres, forgetting an entry also retires the current facts, dream edges and lessons it was the last surviving source of, keeping them as history, and queues affected session digests for regeneration); lesson and world scopes retire the slot with an audit row and are reversible with memory_graph_review(action="restore_slot", store=..., src="entity|attribute"); for "keep the history but mark it wrong" use memory_supersede instead:

memory_forget(scope="memory", source="test-noise")              # refused past 20 matches
memory_forget(scope="memory", source="test-noise", confirm_bulk=True)
memory_forget(scope="fact", entity="test-entity")

Discovering what's in the bank: open the Cortex Console — sources, tags, episodes, and full-table views all live there. Going deeper: reranking, BM25, abstention, and trace debugging · episodes + tags · canonical facts, contenders, world facts, lessons · the consolidation workflow.

Dreaming — consolidating memories into facts

A dream distils the recent associative stream into canonical cortex facts while you're not looking: pull unconsolidated memories → extract (entity, attribute, value) → acknowledge those exact entries durably. New entries remain pending regardless of their timestamps; failed acknowledgement can replay claim application. Manual commits use the token returned by pull. Extraction is pluggable:

The stack ships tier 2 preconfigured (the bespoke Gemma 4 E4B extractor fine-tune in a llama.cpp sidecar, internal-only). The sweep cadence, pointing dreams at a bigger local model or at Claude Sonnet with automatic sidecar fallback, the full-corpus deep dream graph pass, and the privacy/cost trade-offs: Dreaming.

Benchmarks

The headline is the whole benchmark, not a slice: all six LongMemEval question types, 500 questions, oracle variant, run end to end through the memory (qwen-27b extraction under the v25 embedding backbone, BM25-on turn retrieval). Single pass, graded by the local Qwen3.6-27B bench judge (2026-08-03):

The cascade is a serving policy, not a fourth pipeline: answer from the consolidated facts when that channel commits, fall back to raw-turn RAG when it abstains. Overall this is a wash on accuracy at ~73% of the context — 0.690 vs 0.688 is one question in 500 on a single pass, and nobody should read it as a win. The fact spine alone answers at ~13% of RAG's token budget, at a large accuracy cost outside the types it is built for. The structure is per type:

The consolidated spine helps where a fact changes and where the answer sits inside one session; it loses where the answer must be aggregated across sessions or ordered in time, because per-fact consolidation is exactly what discards that structure. BEAM-100K reproduces the same shape independently. Its abstention questions are where the spine looks best — the fact-spine arm scores 0.950 against naive RAG's 0.775, identical under the local judge and under an independent Opus-class judge — but the 2026-09-02 five-arm run bounds that reading: a no-memory arm scores 1.000 there, so the edge is calibration (a small fact context refuses where raw turns confabulate), not evidence that the memory recalled anything. Setup, caveats, both bench stacks side by side, and the evidence that extraction quality is the dominant factor: Benchmarks; full methodology: evals/README.md.

Retrieval itself was re-measured on the same corpus before the v25 backbone swap (150 questions, 74,183 haystack turns, 299 gold turns; pure recall — no reader, no judge): Qwen/Qwen3-Embedding-0.6B reaches R@10 0.809 against bge-base-en-v1.5's 0.742 and the previously-shipped all-MiniLM-L6-v2's 0.572, and beats bge-base head-to-head +32/−12 at k=10 (p=0.004). Artifacts: embedder-recall-shootout-20260727.json, embedder-recall-qwen-vs-bge-20260728.json.

Cortex Console (web UI)

An operator dashboard served by the daemon itself — point a browser at http://127.0.0.1:8765/ui/ (the /health and /mcp endpoints are unchanged; the console is additive). It's a read-mostly instrument panel for seeing and steering the memory a human otherwise can't observe: Observatory (health, per-layer counts, the memory store's capacity meter, dream gauges, consolidation review), Cortex (canonical facts by entity with provenance, version-history ladders, Adopt/Discard for contested slots), World / Lessons / Episodes, Stream (live search with rerank/BM25 toggles and a ranking-trace panel), Recall (multi-hop walks and paths between two entities), Graph (a 3D galaxy of entities with an entity page per star, and a table view), Review (the graph review queue: Accept/Reject merges or — for a source file and its own bare concept, band.py ↔ band — record an implements edge instead of forcing merge-or-dismiss; proposals a background dream has already judged carry a verdict chip — accept/reject/leave with confidence, the model's reason in the tooltip — as a lead, never a decision), Insight (hubs, communities and open questions from the graph digest), and Settings (every safe config.yaml scalar with live-vs-restart badges, diff-preview, and atomic save). Bookmarks into the pre-v3 console (#/console, #/atlas, #/coordination) land on the matching view.

The Board is a read-only coordination view: active roster/status age, reported children, park needs/resume, resource holders and FIFO queues. Pending mail counts are visible only for the caller's principal; the latest 100 retained send/read/ack/wake/expiry events omit all message bodies. This view requires a configured bearer on coordination.allowed_principals, even on an otherwise open loopback installation. Refresh neither receives/acknowledges mail nor settles expired leases. Snapshots are marked stale after a minute; use Refresh for current state. The API is GET /api/agents?view=coordination&limit=50 (roster limit 1–50); the default /api/agents awareness response is unchanged. Incoming audit visibility uses current recipient ow