A structured memory compression system that optimizes context windows by summarizing conversation history into a machine-readable format.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install compact-summary
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 compact-summary 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 Compact Summary?
The Compact Summary skill is a sophisticated memory management tool designed to enhance how AI agents handle long-running conversations. By utilizing a structured format for writing to the MEMORY.md file, this skill ensures that vital context is never lost during token flushes. This approach is one of the most effective Openclaw Skills for developers who need to maintain high-fidelity project state without hitting context window limits.
Inspired by the XML-based summary formats found in professional coding agents, this skill transforms messy chat logs into organized, actionable data. It allows the agent to preserve current work progress, pending tasks, and key file references, making it easier to resume complex workflows after a memory reset.
Compact Summary Use Cases
- Managing large-scale software development tasks that exceed the agent's context window.
- Consolidating multi-turn research sessions into a single source of truth.
- Transitioning between different work sessions while maintaining a persistent todo list and timeline.
- Streamlining the memory flush process to prevent the agent from losing track of current P0 tasks.
How Compact Summary Works
- The system monitors the conversation history and triggers the compaction process during a memory flush.
- The skill analyzes the current scope, including the number of messages and tools invoked such as exec or read_file.
- A structured summary is generated, categorizing data into specific fields like User Requests, Key Files, and Timeline.
- The formatted summary is written directly to the MEMORY.md file in the workspace.
- The agent follows the Resume Instruction to continue work immediately without unnecessary recaps.
Compact Summary Setup
To integrate this memory management flow into your environment, ensure your agent is configured to use Openclaw Skills for context handling.
# Define the memory management protocol in your workspace
touch MEMORY.md
# Configure the agent to trigger the Compact Summary format during flush
openclaw config set memory_compaction_style "structured"
Compact Summary Data Schema & Taxonomy
The skill organizes data using a specific schema to ensure the agent can parse its own memory efficiently:
| Field | Description |
|---|---|
| Conversation Scope | Statistics on user, assistant, and tool message counts. |
| Tools Used | A list of all tools invoked during the session (e.g., grep_search). |
| User Requests | The core objectives defined by the user. |
| Todo Items | A checklist of incomplete tasks and P0 priorities. |
| Key Files | Paths to files that were central to the conversation. |
| Timeline | A chronological log of interactions and tool results. |
| Resume Instruction | Guidelines for the agent on how to proceed after compaction. |
Compact Summary Advanced Features
- XML-inspired structured formatting for improved LLM parsing.
- Automated timeline generation tracking tool results and assistant analysis.
- Statistically driven conversation scope reporting.
- Silent resume protocols to eliminate conversational filler after memory resets.
- Support for multi-agent coordination via shared MEMORY.md files using Openclaw Skills.
SKILL.md
Loading
Related Openclaw Skills
Audit Log Hook
A robust logging mechanism that records every tool interaction for security and performance analysis.
Error Recovery
A comprehensive framework and pattern set for implementing resilient failure handling and state recovery in AI agent workflows.
OpenClaw Hook Examples
A comprehensive developer guide for implementing and utilizing lifecycle hooks to intercept, audit, and modify AI agent behavior within the OpenClaw ecosystem.
Tool Enhancement
A comprehensive suite of over 55 professional-grade tools designed to extend the capabilities of AI agents within the Openclaw Skills ecosystem.
Auto Skill Trigger
An automated pre-filtering mechanism that intelligently matches AI agent skills to specific user tasks based on keyword relevance and pattern scoring.
Sonarr
A powerful integration for managing and automating TV series discovery, library organization, and download monitoring through the Sonarr REST API.
Featured*
