@xdarkicex/openclaw-memory-libravdb 1.4.16 → 1.4.18

package/docs/development.md

@@ -0,0 +1,98 @@
+# Development
+
+This document covers source setup and repository maintenance tasks. For user
+installation, use [Install](./install.md).
+
+## Prerequisites
+
+- Node.js `>= 22`
+- `pnpm`
+- OpenClaw CLI for end-to-end plugin testing
+- a published or locally built `libravdbd` daemon for integration tests
+
+Go is only required when building the daemon from a local daemon checkout or
+regenerating Go gRPC stubs.
+
+## Source Setup
+
+```bash
+pnpm install
+pnpm check
+```
+
+`pnpm check` runs TypeScript validation and unit tests:
+
+```bash
+tsc --noEmit
+tsc -p tsconfig.tests.json && node --test .ts-build/test/unit/*.test.js
+```
+
+## Local Daemon Build
+
+Prepare `.daemon-bin/libravdbd` for local plugin testing:
+
+```bash
+bash scripts/build-daemon.sh
+```
+
+Supported inputs:
+
+- installed daemon on `PATH`, such as `brew install libravdbd`
+- `LIBRAVDBD_BINARY_PATH=/path/to/libravdbd`
+- `LIBRAVDBD_SOURCE_DIR=/path/to/libravdbd` to build from a local daemon repo
+
+For daemon-internal Go development and release work, use the separate
+`libravdbd` repository.
+
+## Validation Commands
+
+```bash
+pnpm check
+npm run test:integration
+```
+
+Benchmark and tuning commands are documented in
+[Performance and tuning](./performance-and-tuning.md).
+
+## Generated IPC Files
+
+The plugin imports generated IPC envelope and RPC payload classes from
+`src/generated/libravdb/ipc/v1/rpc_pb.js`.
+
+Those generated files are checked in and copied into `dist/generated/` during
+`npm run build`:
+
+```bash
+npm run build
+```
+
+Do not replace those imports with the older external
+`@xdarkicex/libravdb-contracts` path. The current package resolves generated
+types from this repository.
+
+## Proto Generation
+
+The repo also contains `api/proto/intelligence_kernel/v1/kernel.proto` and a
+small `Makefile` target for Go gRPC stub generation:
+
+```bash
+make proto
+```
+
+That target assumes Homebrew-style locations for `go`, `protoc`, and the Go
+plugins. Adjust the Makefile locally if your toolchain lives elsewhere.
+
+## Release Shape
+
+The npm package contains:
+
+- `README.md`
+- `HOOK.md`
+- `index.js`
+- `openclaw.plugin.json`
+- `package.json`
+- `docs/`
+- `dist/`
+
+The package is connect-only. It does not compile Go code, download models, or
+manage the daemon process during plugin installation.

package/docs/features.md

@@ -0,0 +1,127 @@
+# Features
+
+This document covers optional plugin features that do not belong in the root
+README: markdown ingestion, Obsidian ingestion, dream promotion, and memory CLI
+commands.
+
+## Markdown Ingestion
+
+LibraVDB Memory can watch markdown roots and sync changed notes into vector
+memory without changing the sidecar RPC contract.
+
+The built-in source adapters are:
+
+- `generic` for OpenClaw-owned markdown, including stock files like `MEMORY.md`
+- `obsidian` for Obsidian vault roots, with tag-aware defaults
+
+Typical usage:
+
+- point `markdownIngestionRoots` at OpenClaw-owned markdown roots, such as
+  `.openclaw/skills/*/*.md` or a directory that contains `MEMORY.md`
+- enable the Obsidian adapter with `markdownIngestionObsidianEnabled: true` and
+  one or more vault roots
+- use include/exclude globs to narrow what gets watched
+
+Example:
+
+```json
+{
+  "plugins": {
+    "configs": {
+      "libravdb-memory": {
+        "markdownIngestionEnabled": true,
+        "markdownIngestionRoots": [
+          "/Users/<you>/.openclaw/memory"
+        ],
+        "markdownIngestionObsidianEnabled": true,
+        "markdownIngestionObsidianRoots": [
+          "/Users/<you>/Documents/Obsidian/Main"
+        ]
+      }
+    }
+  }
+}
+```
+
+Relevant config fields:
+
+| Field | Purpose |
+|---|---|
+| `markdownIngestionEnabled` | Enables or disables generic markdown ingestion. |
+| `markdownIngestionRoots` | Generic markdown roots to watch. |
+| `markdownIngestionInclude` | Optional include globs for generic roots. |
+| `markdownIngestionExclude` | Optional exclude globs for generic roots. |
+| `markdownIngestionCollection` | Target collection for generic markdown, default `global`. |
+| `markdownIngestionDebounceMs` | Watch debounce window, default `150`. |
+| `markdownIngestionObsidianEnabled` | Enables Obsidian ingestion when vault roots exist. |
+| `markdownIngestionObsidianRoots` | Obsidian vault roots to watch. |
+| `markdownIngestionObsidianInclude` | Optional include globs for Obsidian roots. |
+| `markdownIngestionObsidianExclude` | Optional exclude globs for Obsidian roots. |
+| `markdownIngestionObsidianDebounceMs` | Obsidian watch debounce window, default `150`. |
+
+By default, the Obsidian adapter auto-ingests notes that look like memory notes,
+using frontmatter tags or inline tags such as `#project`. The stock OpenClaw
+`MEMORY.md` file is always eligible through the generic adapter path.
+
+## Dream Promotion
+
+Dream promotion is an opt-in path for promoting vetted dream diary entries into
+a dedicated `dream:{userId}` collection.
+
+It does not use `MEMORY.md`. It expects a dream diary markdown file with
+candidate bullets under promotion-oriented headings such as `## Deep Sleep`.
+Each promoted bullet should include trailing metadata with the gating fields:
+
+```md
+- Preserve the recent tail buffer {score=0.82 recall=3 unique=2}
+```
+
+Only bullets that satisfy the sidecar gates are inserted. The dream collection
+is isolated from normal `user:` and `global` retrieval by default, and dream
+phrasing in chat or search queries routes there automatically.
+
+Automatic diary watching:
+
+```json
+{
+  "plugins": {
+    "configs": {
+      "libravdb-memory": {
+        "dreamPromotionEnabled": true,
+        "dreamPromotionDiaryPath": "/Users/<you>/DREAMS.md",
+        "dreamPromotionUserId": "<userId>",
+        "dreamPromotionDebounceMs": 150
+      }
+    }
+  }
+}
+```
+
+Manual run:
+
+```bash
+openclaw memory dream-promote --user-id <userId> --dream-file /path/to/DREAMS.md
+```
+
+The manual command and watcher both use the same sidecar promotion RPC, so
+admission gates and provenance metadata are identical.
+
+## Memory CLI
+
+The plugin registers `openclaw memory` commands when the host exposes the plugin
+CLI API.
+
+| Command | Purpose |
+|---|---|
+| `openclaw memory status` | Show sidecar health, counts, active thresholds, and model readiness. |
+| `openclaw memory index --force` | Refresh delegated sidecar index state for OpenClaw memory CLI compatibility. |
+| `openclaw memory search "query"` | Search LibraVDB memory through the active memory runtime bridge. |
+| `openclaw memory export --user-id <userId>` | Stream stored memories as newline-delimited JSON for one durable namespace. |
+| `openclaw memory export --session-key <sessionKey>` | Export a namespace derived from a session key. |
+| `openclaw memory flush --user-id <userId>` | Delete one durable user namespace after confirmation. |
+| `openclaw memory flush --session-key <sessionKey>` | Delete a namespace derived from a session key after confirmation. |
+| `openclaw memory journal --limit 50` | Inspect bounded lifecycle hints recorded by the sidecar. |
+| `openclaw memory dream-promote --user-id <userId> --dream-file <path>` | Promote vetted dream diary bullets. |
+
+Use `--yes` with `flush` only when you intentionally want to skip the
+confirmation prompt.

package/docs/install.md

@@ -4,6 +4,10 @@ LibraVDB Memory is a connect-only OpenClaw plugin. Install the plugin as a
 normal package, install `libravdbd` separately, and point the plugin at the
 daemon endpoint when you need a non-default location.
 
+OpenClaw compatibility note:
+
+- the plugin is currently verified against OpenClaw `2026.4.23`
+
 For deeper operational detail, use the full
 [installation reference](./installation.md).
 

package/docs/installation.md

@@ -1,123 +1,34 @@
 # Installation Reference
 
-This document is the full installation reference for `@xdarkicex/openclaw-memory-libravdb`. For the short path, use the root [README.md](../README.md).
+This is the full installation reference for
+`@xdarkicex/openclaw-memory-libravdb`. For the shortest path, use
+[install.md](./install.md).
 
 ## System Requirements
 
-| Requirement | Minimum | Recommended | Notes |
-|---|---|---|---|
-| Node.js | `22.0.0` | Latest LTS | Enforced in [`package.json`](../package.json) `engines.node` |
-| OpenClaw | `2026.3.22` | Current stable | Pinned by [`package.json`](../package.json) `peerDependencies.openclaw`; this is the earliest local tag confirmed to expose `definePluginEntry`, `registerContextEngine`, `registerMemoryPromptSection`, and the base plugin API shape this repo uses. Newer hosts may also expose the optional `registerMemoryRuntime` seam, which this plugin now adopts when available |
-| Go | `1.22` | Latest stable | Required only for local daemon development, not for normal plugin install |
-| Disk | about `1 GB` free for default Nomic install | `2 GB+` if provisioning optional T5 and leaving room for DB growth | See Resource Requirements below |
-| RAM | about `512 MB` for embed-only runtime | `1 GB+` if optional T5 summarizer is provisioned | Based on local RSS measurements below |
-| OS | macOS, Linux, Windows | Current stable releases | Unix uses a local socket; Windows uses TCP loopback |
-| Architecture | `arm64`, `x64` | Match published daemon release assets | Current release matrix builds five daemon targets |
+| Requirement | Minimum | Notes |
+|---|---:|---|
+| Node.js | `22.0.0` | Enforced by `package.json` `engines.node`. |
+| OpenClaw | `2026.3.22` | Earliest supported host version for this plugin API shape. |
+| `libravdbd` | published daemon asset | Required for normal runtime. |
+| Go | `1.22` | Required only for local daemon development. |
+| OS | macOS, Linux, Windows | Unix uses a local socket; Windows uses TCP loopback. |
+| Architecture | `arm64`, `x64` | Must match the daemon release asset. |
 
-The published plugin install path is scanner-clean and connect-only. End users should not need Go to install the OpenClaw plugin itself.
+Resource sizing and benchmark data live in
+[Performance and tuning](./performance-and-tuning.md).
 
-## Resource Requirements
+OpenClaw compatibility note:
 
-The numbers in this section are either directly measured from the current local
-build on `2026-03-29` or explicitly labeled as estimates.
+- the plugin is currently verified against OpenClaw `2026.4.23`
 
-### Disk
+## Install Flow
 
-Measured locally from this checkout:
+The published plugin package is connect-only. It installs TypeScript plugin code
+and docs; it does not compile Go code, download model assets, or supervise the
+daemon.
 
-- daemon binary: `7.7M`
-- bundled Nomic model directory: `523M`
-- bundled MiniLM fallback model directory: `87M`
-- optional T5 summarizer directory: `371M`
-- unpacked ONNX Runtime directory on macOS arm64: `44M`
-- ONNX Runtime archive download on macOS arm64: `9.5M`
-
-Practical footprints:
-
-- default quality-first install without optional T5:
-  about `575 MB` (`7.7M + 523M + 44M`)
-- install with optional T5 summarizer:
-  about `946 MB`
-
-Vector payload lower bounds for stored turns, derived from embedding dimension:
-
-- MiniLM `384d`: `384 * 4 = 1536 bytes` per vector
-- Nomic `768d`: `768 * 4 = 3072 bytes` per vector
-
-Estimated lower-bound vector payload for `10,000` stored turns:
-
-- MiniLM: about `15.4 MB`
-- Nomic: about `30.7 MB`
-
-These are lower bounds for vector payload only. Actual on-disk LibraVDB usage is
-higher because text, metadata, collection structure, and index state are stored
-as well.
-
-### Memory
-
-Measured locally on Apple M2, `2026-03-29`, by starting the daemon and reading
-RSS after startup:
-
-- idle RSS with Nomic embedding path loaded and no optional T5 summarizer:
-  about `271,872 KB` (`~266 MB`)
-- idle RSS with Nomic plus local ONNX T5 summarizer loaded:
-  about `515,312 KB` (`~503 MB`)
-
-Not yet bench-measured in the repo:
-
-- RSS during active inference
-- peak RSS during compaction of large clusters
-
-Current operational estimate:
-
-- embedding inference should remain close to the embed-only idle baseline plus
-  transient ONNX workspace allocation
-- optional T5 provisioning roughly doubles steady-state RSS
-
-### CPU
-
-Measured locally from the existing Go benchmark harness on Apple M2,
-`2026-03-29`:
-
-- MiniLM bundled query embedding: about `22.6 ms/op`
-- MiniLM onnx-local query embedding: about `16.3 ms/op`
-- Nomic onnx-local query embedding: about `43.7 ms/op`
-
-Measured locally from a one-off 40-query timing sample on Apple M2,
-`2026-03-29`:
-
-- Nomic query embedding `p50`: about `18.61 ms`
-- Nomic query embedding `p95`: about `24.19 ms`
-
-Measured locally from a one-off synthetic 50-turn compaction run using the
-current extractive summarizer and Nomic embeddings:
-
-- `50`-turn extractive compaction wall time: about `3175 ms`
-
-Not yet bench-measured in the repo:
-
-- equivalent Linux x64 embedding latency on a reference machine
-- `50`-turn compaction wall time through the optional ONNX T5 abstractive path
-
-### Network
-
-Setup downloads are front-loaded. After installation, the plugin is local-first.
-
-Current setup assets:
-
-- Nomic model: about `522 MB`
-- T5-small encoder: about `135 MB`
-- T5-small decoder: about `222 MB`
-- ONNX Runtime macOS arm64 archive: about `9.5 MB`
-
-After install, the plugin makes no required network calls for embedding or
-extractive compaction. The only optional runtime network path is:
-
-- `summarizerBackend = "ollama-local"` or another custom summarizer endpoint
-
-## Standard Install
-
-### Fastest Path on macOS
+Recommended macOS path:
 
 ```bash
 brew tap xDarkicex/homebrew-openclaw-libravdb-memory
@@ -126,85 +37,30 @@ brew services start libravdbd
 openclaw plugins install @xdarkicex/openclaw-memory-libravdb
 ```
 
-This is the preferred install flow for macOS users. It gives you a managed `libravdbd` service and a scanner-clean OpenClaw plugin package.
-
-### Plugin Package
-
-```bash
-openclaw plugins install @xdarkicex/openclaw-memory-libravdb
-```
-
-The plugin package installs as compiled OpenClaw runtime code without daemon bootstrap hooks.
-
-## Daemon Install
-
-Install and start `libravdbd` separately for the same user account that runs OpenClaw. The daemon owns the local DB engine and listens on a local endpoint.
-
-Default endpoints:
-
-- Homebrew on macOS: `unix:/opt/homebrew/var/clawdb/run/libravdb.sock`
-- macOS/Linux user-local installs: `unix:$HOME/.clawdb/run/libravdb.sock`
-- Windows: `tcp:127.0.0.1:37421`
-
-If you run the daemon on a different endpoint, set `plugins.configs.libravdb-memory.sidecarPath` in `~/.openclaw/openclaw.json`.
-
-### Linux
-
-Recommended layout:
+Manual Linux sketch:
 
 ```bash
 mkdir -p ~/.local/bin ~/.config/systemd/user
 curl -L -o ~/.local/bin/libravdbd <published-libravdbd-binary-url>
 chmod +x ~/.local/bin/libravdbd
-cp <published-libravdbd-service-template> ~/.config/systemd/user/libravdbd.service
+curl -L -o ~/.config/systemd/user/libravdbd.service <published-libravdbd-service-template-url>
 systemctl --user enable --now libravdbd.service
+openclaw plugins install @xdarkicex/openclaw-memory-libravdb
 ```
 
-Then verify:
-
-```bash
-systemctl --user status libravdbd.service
-openclaw memory status
-```
-
-### Homebrew / macOS
-
-Homebrew users should normally install from the published tap:
-
-```bash
-brew tap xDarkicex/homebrew-openclaw-libravdb-memory
-brew install libravdbd
-brew services start libravdbd
-```
-
-With `sidecarPath: "auto"`, Homebrew installs on Apple Silicon should resolve to:
-
-```text
-unix:/opt/homebrew/var/clawdb/run/libravdb.sock
-```
-
-User-local installs still default to:
+Windows uses a loopback TCP endpoint by default:
 
 ```text
-unix:$HOME/.clawdb/run/libravdb.sock
+tcp:127.0.0.1:37421
 ```
 
-The daemon release pipeline generates a publish-ready `libravdbd.rb` formula asset for release assets named:
-
-- `libravdbd-darwin-arm64`
-- `libravdbd-darwin-amd64`
-- `libravdbd-linux-amd64`
-- `libravdbd-linux-arm64`
+This repository does not yet include a full Windows service-install walkthrough.
+Use the published Windows daemon asset under your preferred process supervisor
+or run `libravdbd serve` in a terminal for validation.
 
-The generated Homebrew formula also stages the bundled ONNX Runtime archive, the shipped embedding profile assets, and the T5 summarizer bundle into the install prefix so the daemon can start without a separate manual asset unpack step.
-
-If your GitHub Actions configuration includes:
-
-- public tap repository `xDarkicex/homebrew-openclaw-libravdb-memory`
-
-then tagged releases also push the generated formula into `Formula/libravdbd.rb` in that tap repository automatically.
+## Activation
 
-Example plugin config:
+Assign `libravdb-memory` to both OpenClaw slots:
 
 ```json
 {
@@ -212,29 +68,15 @@ Example plugin config:
     "slots": {
       "memory": "libravdb-memory",
       "contextEngine": "libravdb-memory"
-    },
-    "configs": {
-      "libravdb-memory": {
-        "sidecarPath": "unix:/Users/<you>/.clawdb/run/libravdb.sock"
-      }
     }
   }
 }
 ```
 
-## Expected Install Shape
+Treat partial assignment as a misconfiguration. This plugin is designed to own
+memory prompt injection and the context-engine lifecycle together.
 
-Expected successful plugin install shape:
-
-```text
-Installed plugin: libravdb-memory
-```
-
-## Activation
-
-The manifest declares `kind: "context-engine"` and the runtime registers the memory prompt and memory runtime surfaces in code. It is still intended to own both the `memory` and `contextEngine` slots together. Treat partial slot assignment as a misconfiguration.
-
-Add this to `~/.openclaw/openclaw.json`:
+If the daemon uses a non-default endpoint, add `sidecarPath`:
 
 ```json
 {
@@ -242,22 +84,37 @@ Add this to `~/.openclaw/openclaw.json`:
     "slots": {
       "memory": "libravdb-memory",
       "contextEngine": "libravdb-memory"
+    },
+    "configs": {
+      "libravdb-memory": {
+        "sidecarPath": "unix:/Users/<you>/.clawdb/run/libravdb.sock"
+      }
     }
   }
 }
 ```
 
-Notes:
+When `sidecarPath` is `"auto"`, macOS/Linux endpoint resolution checks:
+
+1. `LIBRAVDB_RPC_ENDPOINT`
+2. `$HOME/.clawdb/run/libravdb.sock`
+3. `/opt/homebrew/var/clawdb/run/libravdb.sock`
+4. `/usr/local/var/clawdb/run/libravdb.sock`
+5. fallback to `$HOME/.clawdb/run/libravdb.sock`
 
-- This plugin should own both `memory` and `contextEngine`. Do not assign only one of them.
-- The plugin id is `libravdb-memory`. The npm package name used at install time is `@xdarkicex/openclaw-memory-libravdb`.
-- On newer OpenClaw versions, the plugin also registers a memory runtime bridge so the built-in `memory_search` tool can query libraVDB through the same sidecar-backed retrieval path.
-- On newer OpenClaw versions, the plugin also listens for `before_reset` and `session_end` so it can send best-effort lifecycle hints into the sidecar.
-- Those hints are journaled internally by the sidecar and can be inspected with `openclaw memory journal` without exposing them to normal memory export or recall.
-- The journal keeps only a bounded number of newest entries. Override that cap with `plugins.configs.libravdb-memory.lifecycleJournalMaxEntries` if you need a different retention window.
-- The plugin does not currently register `registerMemoryFlushPlan`; transcript ingest and compaction remain owned by the context-engine lifecycle and the sidecar.
+## Default Paths
 
-Without a slot entry, OpenClaw's default memory can continue to run in parallel.
+| Platform | Default endpoint |
+|---|---|
+| macOS/Linux user-local | `unix:$HOME/.clawdb/run/libravdb.sock` |
+| macOS Homebrew Apple Silicon | `unix:/opt/homebrew/var/clawdb/run/libravdb.sock` |
+| Windows | `tcp:127.0.0.1:37421` |
+
+Default data path:
+
+```text
+$HOME/.clawdb/data.libravdb
+```
 
 ## Verification
 
@@ -274,6 +131,7 @@ Expected output shape:
 │ Sidecar            │ running                      │
 │ Turns stored       │ 0                            │
 │ Memories stored    │ 0                            │
+│ Lifecycle hints    │ 0                            │
 │ Gate threshold     │ 0.35                         │
 │ Abstractive model  │ ready | not provisioned      │
 │ Embedding profile  │ all-minilm-l6-v2             │
@@ -283,51 +141,10 @@ Expected output shape:
 
 Interpretation:
 
-- `Sidecar=running` means the local `libravdbd` daemon answered JSON-RPC `health`.
-- `Gate threshold=0.35` confirms the default gating scalar boundary is active.
-- `Abstractive model=not provisioned` is acceptable. The system degrades to extractive compaction.
-
-## Contributor Install
-
-For contributors working from a clone:
-
-```bash
-pnpm check
-bash scripts/build-daemon.sh
-```
-
-This prepares a local daemon binary in `.daemon-bin/libravdbd` (or `.exe` on Windows) and copies any locally available model/runtime assets there for testing.
-
-Contributor default:
-
-- install `libravdbd` separately with Homebrew or release assets, then run `bash scripts/build-daemon.sh`
-
-Private local daemon development:
-
-- set `LIBRAVDBD_SOURCE_DIR=/path/to/libravdbd` to build from your local daemon repo
-- or set `LIBRAVDBD_BINARY_PATH=/path/to/libravdbd` to use a prebuilt local daemon binary
-
-## User-Service Templates
-
-Published daemon installs include matching user-service templates:
-
-- Linux user service: `libravdbd.service`
-- macOS LaunchAgent: `com.xdarkicex.libravdbd.plist`
-
-Linux example:
-
-```bash
-mkdir -p ~/.config/systemd/user
-cp <published-libravdbd-service-template> ~/.config/systemd/user/libravdbd.service
-systemctl --user enable --now libravdbd.service
-```
-
-macOS example:
-
-1. Copy the published `com.xdarkicex.libravdbd.plist`
-2. Replace `__LIBRAVDBD_PATH__` and `__HOME__`
-3. Save it to `~/Library/LaunchAgents/com.xdarkicex.libravdbd.plist`
-4. Load it with `launchctl load ~/Library/LaunchAgents/com.xdarkicex.libravdbd.plist`
+- `Sidecar=running` means the daemon answered the health check.
+- `Gate threshold=0.35` confirms the default durable-memory gate.
+- `Abstractive model=not provisioned` is acceptable; compaction falls back to
+  the extractive path.
 
 ## Troubleshooting
 
@@ -335,47 +152,37 @@ macOS example:
 
 Common causes:
 
-- ONNX Runtime library missing or unpacked in the wrong place
-- downloaded model file hash mismatch
-- `libravdbd` not started for the current user
-- plugin pointed at the wrong endpoint
+- `libravdbd` is not running for the same user account as OpenClaw
+- `sidecarPath` points at the wrong endpoint
+- ONNX Runtime assets are missing or unpacked in the wrong place
+- a model asset failed checksum validation
 
-Check:
+Check the daemon first:
 
 ```bash
 openclaw memory status
+brew services restart libravdbd
 ```
 
-If the daemon is down, start it and verify the configured endpoint:
-
-```bash
-brew services start libravdbd
-```
-
-Or, without Homebrew:
+For foreground debugging:
 
 ```bash
 libravdbd serve
 ```
 
-On macOS/Linux, the default endpoint is `unix:$HOME/.clawdb/run/libravdb.sock`. On Windows, the default endpoint is `tcp:127.0.0.1:37421`.
-
 ### Hash mismatch
 
-Hash mismatch means one of:
-
-- the daemon asset is corrupt
-- the local cache is stale
-- the expected checksum is wrong
-
-Do not bypass this. Delete the asset and rerun setup, or republish the release with corrected checksums.
+Do not bypass a checksum mismatch. Delete the corrupt or stale asset and rerun
+setup, or republish the release with corrected checksums.
 
-### Windows behavior
+### Default memory still appears active
 
-On Windows the daemon uses a loopback TCP endpoint instead of a Unix socket. This is expected. The plugin’s transport layer already handles the fallback.
+Confirm that `libravdb-memory` is assigned to both `memory` and
+`contextEngine`. Without both slot entries, OpenClaw's default memory path can
+continue to run in parallel.
 
-### Published daemon requirement
+### Lifecycle journal looks empty
 
-The daemon must come from a published `libravdbd` binary for the current platform.
-If that download or checksum verification fails, setup stops instead of falling
-back to a local `go build`.
+The sidecar journal only records advisory lifecycle hints such as `before_reset`
+and `session_end`. It is bounded by `lifecycleJournalMaxEntries`, default `500`,
+and is not part of normal memory recall.