diff --git a/docs/cli/memory.md b/docs/cli/memory.md
index 3765f670c17..3b198ee1432 100644
--- a/docs/cli/memory.md
+++ b/docs/cli/memory.md
@@ -34,6 +34,8 @@ openclaw memory status --deep --index
 openclaw memory status --deep --index --verbose
 openclaw memory status --agent main
 openclaw memory index --agent main --verbose
+openclaw memory promote-explain "router vlan"
+openclaw memory rem-harness --json
 ```
 
 ## Options
@@ -90,6 +92,33 @@ Full options:
 - `--include-promoted`: include already promoted candidates in output.
 - `--json`: print JSON output.
 
+`memory promote-explain`:
+
+Explain why a specific candidate would or would not promote, with a full score breakdown.
+
+```bash
+openclaw memory promote-explain "<selector>"
+```
+
+- `<selector>`: candidate key, path fragment, or snippet fragment to match.
+- `--agent <id>`: scope to a single agent (default: the default agent).
+- `--include-promoted`: include already promoted candidates.
+- `--json`: print JSON output.
+
+`memory rem-harness`:
+
+Preview REM reflections, candidate truths, and deep promotion output without writing anything. Useful for staging and debugging the REM phase before it runs for real.
+
+```bash
+openclaw memory rem-harness [--json] [--agent <id>] [--include-promoted]
+```
+
+- `--agent <id>`: scope to a single agent (default: the default agent).
+- `--include-promoted`: include already promoted deep candidates.
+- `--json`: print JSON output.
+
+See [Dreaming](/concepts/dreaming) for the full phase model and how REM fits in.
+
 ## Dreaming (experimental)
 
 Dreaming is the background memory consolidation system with three cooperative
diff --git a/docs/concepts/dreaming.md b/docs/concepts/dreaming.md
index 9730f430d69..7062028345b 100644
--- a/docs/concepts/dreaming.md
+++ b/docs/concepts/dreaming.md
@@ -225,6 +225,20 @@ openclaw memory status --deep
 
 See [memory CLI](/cli/memory) for the full flag reference.
 
+### Preview and explain tools
+
+Two additional subcommands help you inspect promotion and REM behavior without writing anything:
+
+```bash
+# Explain why a candidate would or would not promote
+openclaw memory promote-explain "meeting notes"
+
+# Preview REM reflections, candidate truths, and deep promotions
+openclaw memory rem-harness --json
+```
+
+See [memory CLI](/cli/memory) for full options.
+
 ## How it works
 
 ### Light phase pipeline
diff --git a/docs/tools/lobster.md b/docs/tools/lobster.md
index 377b37190da..ee3e235c726 100644
--- a/docs/tools/lobster.md
+++ b/docs/tools/lobster.md
@@ -36,7 +36,7 @@ Lobster is intentionally small. The goal is not "a new language," it's a predict
 
 ## How it works
 
-OpenClaw launches the local `lobster` CLI in **tool mode** and parses a JSON envelope from stdout.
+OpenClaw runs Lobster workflows **in-process** using an embedded runner. No external CLI subprocess is spawned; the workflow engine executes inside the gateway process and returns a JSON envelope directly.
 If the pipeline pauses for approval, the tool returns a `resumeToken` so you can continue later.
 
 ## Pattern: small CLI + JSON pipes + approvals
@@ -155,7 +155,9 @@ Notes:
 
 ## Install Lobster
 
-Install the Lobster CLI on the **same host** that runs the OpenClaw Gateway (see the [Lobster repo](https://github.com/openclaw/lobster)), and ensure `lobster` is on `PATH`.
+Bundled Lobster workflows run in-process; no separate `lobster` binary is required. The embedded runner ships with the Lobster plugin.
+
+If you need the standalone Lobster CLI for development or external pipelines, install it from the [Lobster repo](https://github.com/openclaw/lobster) and ensure `lobster` is on `PATH`.
 
 ## Enable the tool
 
@@ -287,9 +289,9 @@ Continue a halted workflow after approval.
 
 ### Optional inputs
 
-- `cwd`: Relative working directory for the pipeline (must stay within the current process working directory).
-- `timeoutMs`: Kill the subprocess if it exceeds this duration (default: 20000).
-- `maxStdoutBytes`: Kill the subprocess if stdout exceeds this size (default: 512000).
+- `cwd`: Relative working directory for the pipeline (must stay within the gateway working directory).
+- `timeoutMs`: Abort the workflow if it exceeds this duration (default: 20000).
+- `maxStdoutBytes`: Abort the workflow if output exceeds this size (default: 512000).
 - `argsJson`: JSON string passed to `lobster run --args-json` (workflow files only).
 
 ## Output envelope
@@ -317,17 +319,17 @@ OpenProse pairs well with Lobster: use `/prose` to orchestrate multi-agent prep,
 
 ## Safety
 
-- **Local subprocess only** — no network calls from the plugin itself.
+- **Local in-process only** — workflows execute inside the gateway process; no network calls from the plugin itself.
 - **No secrets** — Lobster doesn't manage OAuth; it calls OpenClaw tools that do.
 - **Sandbox-aware** — disabled when the tool context is sandboxed.
-- **Hardened** — fixed executable name (`lobster`) on `PATH`; timeouts and output caps enforced.
+- **Hardened** — timeouts and output caps enforced by the embedded runner.
 
 ## Troubleshooting
 
-- **`lobster subprocess timed out`** → increase `timeoutMs`, or split a long pipeline.
+- **`lobster timed out`** → increase `timeoutMs`, or split a long pipeline.
 - **`lobster output exceeded maxStdoutBytes`** → raise `maxStdoutBytes` or reduce output size.
 - **`lobster returned invalid JSON`** → ensure the pipeline runs in tool mode and prints only JSON.
-- **`lobster failed (code …)`** → run the same pipeline in a terminal to inspect stderr.
+- **`lobster failed`** → check gateway logs for the embedded runner error details.
 
 ## Learn more