Openclaw Skills memory-compress-free turns long daily agent logs into compact, structured memory summaries with local Node.js processing.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install memory-compress-free
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install memory-compress-free using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is memory-compress-free?
Openclaw Skills memory-compress-free is a lightweight, free log compression skill for AI agents that converts long Markdown daily logs into concise, structured summaries. It is designed for memory archiving workflows where agents need to preserve key events, decisions, and next steps without carrying forward excessive context.
This skill emphasizes simple local execution, single-file processing, and predictable output structure. It uses keyword extraction plus classic compression rules to reduce log volume, making it easier to append distilled memory into a central MEMORY.md file while keeping important context accessible.
memory-compress-free Use Cases
- Compressing a single day of agent activity into a smaller archival summary
- Turning verbose development logs into structured memory entries
- Extracting key events, progress updates, and pending tasks from Markdown logs
- Appending compressed daily summaries into a long-term
MEMORY.mdknowledge file - Running local, no-API-key log reduction for AI agent workflows
- Reviewing whether a log is too repetitive or already concise enough to compress
How memory-compress-free Works
- Prepare a daily log file in Markdown format, typically stored as
memory/YYYY-MM-DD.md. - Invoke the local Node.js compression script through the AI agent using the input log path and an optional output path.
- The script scans the log for important keyword patterns, major progress points, and to-do items.
- It applies structured compression rules inspired by classic writing principles: remove repetition, keep turning points, remove process noise, and retain white space for readability.
- The result is written as a summarized Markdown output file, usually a compressed memory artifact.
- Review the compressed summary and append it to
MEMORY.mdfor long-term retention. - If needed, manually inspect edge cases where key information may not match the keyword-based extraction rules.
memory-compress-free Setup
- Ensure your AI agent environment supports SKILL.md-style skills and can run local commands.
- Install Node.js 16 or later on Windows, macOS, or Linux.
- Place the provided skill files in the agent’s skills directory so the local script is available.
- Make sure your daily logs are stored as Markdown files under
memory/YYYY-MM-DD.md. - Run the compression workflow from the agent with the input log and output destination.
node <skill-directory>/memory-compress.js memory/2026-07-18.md /tmp/compressed.md
- Append the result to your long-term memory file after reviewing the summary.
cat /tmp/compressed.md >> MEMORY.md
- No external API key is required because processing is handled locally by Node.js.
memory-compress-free Data Schema & Taxonomy
The skill is organized around a small, file-based memory pipeline rather than a database or remote API.
| Artifact | Purpose | Format | Notes |
|---|---|---|---|
memory/YYYY-MM-DD.md |
Raw daily agent log input | Markdown | Primary source file for compression |
compressed.md or /tmp/compressed.md |
Generated summary output | Markdown | Structured compressed memory artifact |
MEMORY.md |
Long-term memory archive | Markdown | Appended after summary review |
options |
Optional configuration | Object | Controls mode, format preference, and similar behavior |
callback_url |
Optional async notification target | String | Present in the skill metadata, though local processing is the main workflow |
Metadata taxonomy used by Openclaw Skills for this skill:
slug:memory-compress-freeversion:1.0.0pricing_tier:freecategory:Agentstags: memory, compress, automation, productivity, 记忆管理, 上下文, AI, node, tmp
Core input parameters:
input— required string containing the log data or commandoptions— optional object for mode and formatting preferencescallback_url— optional URL for asynchronous completion notice
Output characteristics:
- Structured summary with key events and pending items
- Preservation of date context for daily archival
- Designed for single-file compression only
memory-compress-free Advanced Features
- Local-only Node.js execution with no external API key requirement
- Keyword-matching extraction for major events and action items
- Classic compression rules that remove repetition while preserving important signal
- Markdown-native workflow that fits AI agent memory archives
- Manual append pattern for long-term
MEMORY.mdconsolidation - Built for single-file daily log compression in Openclaw Skills workflows
- Compatible with multi-platform agent environments on Windows, macOS, and Linux
- Clear upgrade path to paid-tier features such as batch compression, heartbeat automation, and mixed extraction engines
SKILL.md
Loading
Related Openclaw Skills
Memory Manager Free
Openclaw Skills for building a private, local long-term memory system that categorizes, indexes, and searches AI agent memory without external services.
moltbook-filter-free
Openclaw Skills for moltbook-filter-free help you scan, detect, and remove community-feed spam at the client side with configurable rules and JSON feed filtering.
News Aggregator (Free)
Openclaw Skills for news aggregation, filtering, deduplication, and structured summarization across domestic and international tech, military, and social news.
PPT Toolkit Free
Openclaw Skills for PowerPoint creation, editing, template-based generation, and content/visual QA in a lightweight free AI skill.
Markdown 工具箱（免费版）
Openclaw Skills for generating clean, portable Markdown that stays consistent across platforms and editors.
LLM Assistant Hub Free
Openclaw Skills for long-document reasoning that uses layered analysis and assumption detection to produce clearer, safer, structured outputs.
Featured*
