← Back to Skills Marketplace
fuuuuujichaos

Better Memory

by fuuuuujichaos · GitHub ↗ · v1.0.4 · MIT-0
cross-platform ✓ Security Clean
192
Downloads
0
Stars
0
Active Installs
4
Versions
Install in OpenClaw
/install better-memory
Description
One-shot smart installation and ongoing maintenance for a native OpenClaw memory stack with L1 daily logs (`memory/YYYY-MM-DD.md`), L2 sidecar summaries, and...
README (SKILL.md)

Better Memory V2

Smart Install (Recommended)

Run once to inject memory capability additively and generate migration + schedule artifacts:

python3 skills/better-memory/scripts/smart_install.py --workspace .

Supported install customization:

  • --daily-time HH:MM (default 22:00)
  • --weekly-day MON..SUN (default SUN)
  • --weekly-time HH:MM (default 20:00)
  • --monthly-day 1..31 (default 1)
  • --monthly-time HH:MM (default 10:00)
  • --entry-threshold N (default 8)

Memory Layers

  • L1 (native): memory/YYYY-MM-DD.md
  • L2 (sidecar): .openclaw-memory-os/l2/{experience,value,standard}.md
  • L3 (native): managed block in MEMORY.md

Typed taxonomy:

  • experience: think, say, do
  • value: good, bad
  • standard: right, wrong

Commands

  • Capture L1 typed memory: python3 skills/better-memory/scripts/capture_memory.py --workspace . --kind \x3Cexperience|value|standard> --axis \x3Caxis> --topic "\x3Ctopic>" --statement "\x3Cstatement>"
  • Daily review (L1 -> L2): python3 skills/better-memory/scripts/run_daily_review.py --workspace .
  • Weekly rollup (L2 -> L3): python3 skills/better-memory/scripts/run_weekly_rollup.py --workspace .
  • Monthly advisory review (no auto cleanup): python3 skills/better-memory/scripts/run_monthly_review.py --workspace .
  • Apply approved monthly cleanup actions: python3 skills/better-memory/scripts/apply_monthly_cleanup.py --workspace . --rerollup
    • This command also writes an apply summary in .openclaw-memory-os/reviews/apply-summary-*.md.
  • Emit cron template: python3 skills/better-memory/scripts/emit_cron_template.py --workspace .
  • Legacy migration plan prepare/apply:
    • python3 skills/better-memory/scripts/promote_legacy_memory.py --workspace . --prepare
    • python3 skills/better-memory/scripts/promote_legacy_memory.py --workspace . --apply

Compatibility command:

  • python3 skills/better-memory/scripts/refine_memory.py --workspace .
    • Runs daily review then weekly rollup in one command.

Installation Policy

  • Preserve existing AGENTS.md, MEMORY.md, and daily logs.
  • Write only managed blocks in AGENTS.md, MEMORY.md, and optional HEARTBEAT.md.
  • Keep sidecar artifacts outside native memory/ so memory_search behavior stays clean.
  • Generate migration review and editable migration plan; do not auto-import legacy lines unless enabled.

Monthly Cleanup Policy

  • Monthly script only produces a report and suggested actions.
  • Monthly script also produces a disabled-by-default cleanup plan JSON.
  • Do not auto-delete or auto-rewrite memory on monthly review.
  • Resolve conflicts and bloat with user confirmation, enable selected plan actions, then apply.

References

Read references/memory-schema.md for schema and lifecycle details.

Usage Guidance
This skill is coherent and appears to do what it claims, but it is file-modifying: it will create/modify .openclaw-memory-os, write managed blocks in AGENTS.md/MEMORY.md/HEARTBEAT.md, and append entries to memory/YYYY-MM-DD.md. Before installing or running smart_install.py: 1) Back up or commit your workspace (git) so you can review/revert changes. 2) Inspect the generated migration plan and cron-template before enabling any scheduled jobs. 3) Run smart_install and other scripts in a safe/test workspace first if you have important existing memory files. 4) The tool is local-only (no network/credentials), but it will persist changes to your repository — review apply-summary-* and plan JSON files before applying monthly cleanup actions.
Capability Analysis
Type: OpenClaw Skill Name: better-memory Version: 1.0.4 The 'better-memory' skill bundle is a well-structured framework for managing agent memory through hierarchical distillation (L1 raw logs, L2 summaries, and L3 long-term memory). The Python scripts (e.g., bootstrap_memory.py, refine_memory.py) use standard libraries to perform file I/O and text processing within the workspace, maintaining state in a local '.openclaw-memory-os' directory. There are no indicators of data exfiltration, unauthorized network access, or malicious prompt injection; the instructions provided to the agent in SKILL.md and the managed blocks for AGENTS.md/HEARTBEAT.md are strictly limited to memory maintenance tasks.
Capability Assessment
✓ Purpose & Capability
Name/description (better memory, L1/L2/L3 workflows) match the included scripts. Required binaries (python3) are appropriate. The scripts only touch workspace paths (memory/, MEMORY.md, AGENTS.md, HEARTBEAT.md) and a sidecar directory (.openclaw-memory-os), which is consistent with the stated purpose.
✓ Instruction Scope
SKILL.md and the scripts instruct the agent to run local Python commands that create/modify workspace files and sidecar artifacts. The runtime instructions and code only reference local filesystem paths and settings; they do not read unrelated system files, environment variables, or send data to external endpoints.
✓ Install Mechanism
No install spec — this is instruction/code-only and requires python3 on PATH. All code is included in the package (no external downloads or package installs), so there is no hidden remote install step.
✓ Credentials
No environment variables, credentials, or config paths are requested. The scripts save state and settings under .openclaw-memory-os and modify workspace files only, which is proportionate to the skill's stated functionality.
ℹ Persistence & Privilege
The skill writes persistent files in the workspace and a sidecar directory and inserts managed blocks into AGENTS.md, MEMORY.md, HEARTBEAT.md during bootstrap; it also provides an uninstall script to remove those managed blocks and (optionally) the sidecar directory. 'always' is false and model invocation is not disabled — the skill does not demand elevated platform privileges beyond typical skill behavior.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install better-memory
  3. After installation, invoke the skill by name or use /better-memory
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.4
Major upgrade: Switch to a native Python-based memory management system with smart install, typed multi-layer memory, and improved review/cleanup workflows. - Added scripts for smart installation, daily/weekly/monthly memory review, legacy migration, and cron template generation. - Introduced structured memory layers (L1 daily logs, L2 sidecar summaries, L3 rollups) and typed taxonomy (`experience`, `value`, `standard`). - Moved configuration and operations from JavaScript to Python; removed legacy JS files and configs. - Improved migration safety: preserves existing user data, generates reviews and editable plans instead of automatic changes. - Monthly review now produces non-destructive reports and actions for user-approved cleanup. - Updated documentation to detail new commands, installation options, and compatibility guidance.
v1.0.3
- Added automation instructions for scheduling `runDistiller` and `runMetabolism` using OpenClaw's cron tool. - Provided example cron commands for daily execution of memory tasks. - Included guidance for integrating memory tasks into `HEARTBEAT.md` for automated check-ins. - Added a new README.md file for improved documentation.
v1.0.2
- README.md file removed. - No changes to core logic or features.
v1.0.1
- Initial release of better-memory skill for OpenClaw. - Introduces a hierarchical, self-evolving memory system with four abstraction levels (L0–L3). - Adds an automated metabolism system for garbage collection and conflict resolution. - Provides configurable settings for memory limits and distillation intervals via config.json. - Exposes core API functions: log_event, run_distiller, and run_metabolism.
Metadata
Slug better-memory
Version 1.0.4
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 4
Frequently Asked Questions

What is Better Memory?

One-shot smart installation and ongoing maintenance for a native OpenClaw memory stack with L1 daily logs (`memory/YYYY-MM-DD.md`), L2 sidecar summaries, and... It is an AI Agent Skill for Claude Code / OpenClaw, with 192 downloads so far.

How do I install Better Memory?

Run "/install better-memory" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Better Memory free?

Yes, Better Memory is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Better Memory support?

Better Memory is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Better Memory?

It is built and maintained by fuuuuujichaos (@fuuuuujichaos); the current version is v1.0.4.

More Skills
self-improving agent
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Clau...
⬇ 463283 · by pskoett
Skill Vetter
Security-first skill vetting for AI agents. Use before installing any skill from ClawdHub, GitHub, or other sources. Checks for red flags, permission scope, and suspicious patterns.
⬇ 259410 · by spclaudehome
Polymarket
Query Polymarket prediction markets. Check odds, find trending markets, search events, track price movements.
⬇ 232503 · by joelchance
Self-Improving + Proactive Agent
Self-reflection + Self-criticism + Self-learning + Self-organizing memory. Agent evaluates its own work, catches mistakes, and improves permanently. Use when...
⬇ 200423 · by ivangdavila
Github
Interact with GitHub using the `gh` CLI. Use `gh issue`, `gh pr`, `gh run`, and `gh api` for issues, PRs, CI runs, and advanced queries.
⬇ 191115 · by steipete
ontology
Typed knowledge graph for structured agent memory and composable skills. Use when creating/querying entities (Person, Project, Task, Event, Document), linkin...
⬇ 190156 · by oswalpalash

💬 Comments