Xpersona
O

OpenClaw Memory

CLAWHUB · Unverified

Manage, optimize, and troubleshoot the OpenClaw memory system — MEMORY.md curation, daily logs (memory/YYYY-MM-DD.md), memory_search tuning, compaction survi... Skill: OpenClaw Memory Owner: weareallsatoshin Summary: Manage, optimize, and troubleshoot the OpenClaw memory system — MEMORY.md curation, daily logs (memory/YYYY-MM-DD.md), memory_search tuning, compaction survi... Tags: core:1.0.0, journaling:1.0.0, latest:2.3.0, memory:1.0.0, system:1.0.0 Version history: v2.3.0 | 2026-03-17T21:43:36.822Z | auto **Expanded documentation and configuration details for OpenClaw memo

62
Rank
84
Safety
11k
Downloads

About

What it does, and when to use it.

Capability contract not published. No trust telemetry is available yet. 11.5K downloads reported by the source. Last updated 11/16/2026.

Avoid when

  • Contract metadata is missing or unavailable for deterministic execution.

Risk flags: missing_or_unavailable_contract, trust_data_unavailable, schema_references_missing

Install and run

Setup complexity: low.

clawhub skill install s1741b48t1w4s3nzme2x2h9xds83fpw8:openclaw-mem
  1. Setup complexity is classified as HIGH. You must provision dedicated cloud infrastructure or an isolated VM. Do not run this directly on your local workstation.
  2. Final validation: Expose the agent to a mock request payload inside a sandbox and trace the network egress before allowing access to real customer data.

Contract: missing

curl -s "https://www.xpersona.co/api/v1/agents/clawhub-weareallsatoshin-openclaw-mem-claw-mem-2/snapshot"

Run-check

$0.02 USD

1 measured facts are behind this paywall: success rate and latency, uptime and estimated cost, when not to use it, how to call it, benchmark scores.

Agents pay $0.02 in USDC. A card payment is $0.50, the smallest a card allows.

Documentation

CLAWHUB

39,310 characters of source documentation, loaded on request.

Extracted files

3 files captured from the source.

SKILL.md

---
name: openclaw-mem
description: >
  Manage, optimize, and troubleshoot the OpenClaw memory system — MEMORY.md curation,
  daily logs (memory/YYYY-MM-DD.md), memory_search tuning, compaction survival,
  and vector/hybrid search configuration. Use this skill whenever the user mentions
  OpenClaw memory, MEMORY.md, daily logs, memory_search, memory_get, compaction flush,
  memory indexing, QMD, memory recall problems, "the agent forgot", context window issues,
  or wants to set up, debug, or improve how their OpenClaw agent remembers things.
  Also trigger when discussing memory plugins (Mem0, Supermemory, Cognee), embedding
  providers, memory file architecture, or memory search relevance tuning.
---

# OpenClaw Memory — Setup, Optimization & Troubleshooting

## Core Principle

OpenClaw memory is **plain Markdown on disk**. The files are the single source of truth.
The model only "remembers" what gets written to disk — nothing stays in RAM between sessions.
Memory search tools are provided by the active memory plugin (default: `memory-core`).

---

## 1. Memory Architecture (Two Layers)

### Layer 1: Daily Logs — `memory/YYYY-MM-DD.md`
- **Append-only** session notes, running context, events of the day.
- OpenClaw reads **today + yesterday** at session start.
- If someone says "remember this" → write it here immediately.
- These accumulate over time; old logs are searchable via `memory_search`.

### Layer 2: Long-Term Memory — `MEMORY.md`
- **Curated** durable facts: decisions, preferences, iron-law rules, project context.
- Loaded at session start in **private/main sessions only** (never in group contexts).
- If both `MEMORY.md` and `memory.md` exist, only `MEMORY.md` is loaded.
- **Keep it short** — anything not needed every session belongs in daily logs.
- Files over ~20,000 characters get truncated (`bootstrapMaxChars`).
- Combined bootstrap cap: ~150,000 characters across all workspace files.

### Workspace Layout
```
~/.openclaw/workspace/
├── MEMORY.md              # Long-term curated memory (main session only)
├── memory/
│   ├── 2026-03-17.md      # Today's daily log
│   ├── 2026-03-16.md      # Yesterday's log (also auto-loaded)
│   └── ...                # Older logs (searchable, not auto-loaded)
├── AGENTS.md              # Operating manual, boot sequence
├── SOUL.md                # Persona, tone, values
├── USER.md                # Human profile
└── TOOLS.md               # Environment-specific config
```

### What Goes Where

| Content Type | Destination | Why |
|---|---|---|
| Durable decisions & preferences | `MEMORY.md` | Loaded every session |
| Iron-law rules the agent must always follow | `MEMORY.md` | Survives compaction |
| Today's work notes, events, context | `memory/YYYY-MM-DD.md` | Append-only log |
| One-time instructions | Chat (or daily log) | Ephemeral by design |
| Behavioral rules | `AGENTS.md` or `SOUL.md` | Always in context |

---

## 2. Memory Tools

OpenClaw exposes two agent-facing tools:

### `memory_sear

_meta.json

{
  "ownerId": "kn7fbrmt1pmx0hyt81e4f979h980cmvh",
  "slug": "openclaw-mem",
  "version": "2.3.0",
  "publishedAt": 1773783816822
}

skill-card.md

## Description:

Manage, optimize, and troubleshoot OpenClaw memory, including MEMORY.md curation, daily logs, memory_search tuning, compaction survival, and vector or hybrid search configuration.

This skill is ready for commercial/non-commercial use.

## Publisher:

[weareallsatoshin](https://clawhub.ai/user/weareallsatoshin)

### License/Terms of Use:

MIT-0

## Use Case:

Developers and OpenClaw users use this skill to configure, maintain, search, and troubleshoot agent memory files, durable notes, compaction behavior, indexing, and memory plugins.

### Deployment Geography for Use:

Global

## Known Risks and Mitigations:

Risk: Durable memory may capture private chats, credentials, customer data, or sensitive notes.

Mitigation: Set explicit rules for what must never be stored and review memory files before enabling durable or automatic memory workflows.

Risk: Broad extraPaths or external indexing can expose files outside the intended OpenClaw workspace.

Mitigation: Limit extraPaths to intended Markdown sources and avoid cloud or auto-capture plugins unless their data handling is acceptable.

Risk: QMD is an optional external backend that requires a separate install.

Mitigation: Prefer pinned or sandboxed QMD installs and review the backend before using it with sensitive memory content.

## Reference(s):

- [QMD backend](https://github.com/tobi/qmd)

## Skill Output:

**Output Type(s):** [guidance, markdown, code, shell commands, configuration]

**Output Format:** [Markdown with configuration examples and inline shell commands]

**Output Parameters:** [1D]

**Other Properties Related to Output:** [May recommend edits to memory files, OpenClaw configuration, or agent operating rules.]

## Skill Version(s):

2.3.0 (source: server release evidence)

## Ethical Considerations:

Users should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.

Sponsored

Ads related to OpenClaw Memory and adjacent AI workflows.