← 返回 Skills 市场
Compact Summary
作者
hanxiao-bot
· GitHub ↗
· v1.0.0
· MIT-0
127
总下载
0
收藏
1
当前安装
1
版本数
在 OpenClaw 中安装
/install compact-summary
功能描述
Generates structured, concise memory summaries during flush or compaction, detailing conversation stats, tools, user requests, tasks, files, progress, and ti...
使用说明 (SKILL.md)
Compact Summary - Structured Memory Compression
Purpose
Improve OpenClaw's memory flush and compaction flow by using a structured format for writing to MEMORY.md.
Structured Summary Format
Inspired by the \x3Csummary> XML format from claw-code, the improved format:
# 2026-04-01 Memory Summary
## Conversation Scope
- Original message count: 12 (user=4, assistant=5, tool=3)
- Compression time: 2026-04-01 23:00
## Tools Used
- exec, read_file, grep_search
## User Requests
- Research Claw Code architecture
- Analyze MCP orchestration system
- Compare Tool system differences
## Todo Items
- [ ] Compile improvement list
- [ ] Begin P0 execution
## Key Files
- $WORKSPACE/claw-code/rust/crates/runtime/src/compact.rs
- $WORKSPACE/claw-code/rust/crates/tools/src/lib.rs
## Current Work
Continue executing P0 tasks
## Timeline
- user: Research Claw Code architecture
- assistant: Started analysis
- tool: grep_search pattern=COMPACT
- tool_result: 15 matches
- user: Analyze MCP orchestration
- assistant: MCP has process pool management
## Resume Instruction
Continue directly — do not acknowledge this summary, do not recap what was happening, and do not preface with continuation text.
Trigger Condition
Use this format during memory flush or compaction.
Key Fields
| Field | Description |
|---|---|
| Conversation Scope | Original message statistics |
| Tools Used | Tools actually invoked |
| User Requests | Main user needs |
| Todo Items | Incomplete tasks |
| Key Files | Files involved |
| Current Work | Ongoing tasks |
| Timeline | Conversation summary |
安全使用建议
This skill is instruction-only and generally coherent with its purpose, but review two things before enabling it: (1) filesystem access — the template references $WORKSPACE and concrete file paths, so ensure the agent is permitted to read only appropriate directories and that no secrets reside there; (2) the resume instruction — it tells the agent to continue without acknowledging the summary, which can silently change behavior; if you need explicit user-facing continuity, modify the SKILL.md to require an explicit confirmation or to declare the WORKSPACE env var and permitted file patterns. If you want tighter safety, run it in a sandboxed agent or restrict the agent's file-read and tool-use permissions (especially for exec/grep-style tools).
功能分析
Type: OpenClaw Skill
Name: compact-summary
Version: 1.0.0
The skill bundle provides a structured template for summarizing conversation history into a MEMORY.md file to optimize agent memory. It contains no executable code, network activity, or data exfiltration logic, and the instructions in SKILL.md are strictly aligned with the stated purpose of memory compaction.
能力评估
Purpose & Capability
Skill name/description (compact/summary for memory flush/compaction) aligns with the SKILL.md template: it formats conversation stats, tools used, todo items, key files, and timeline. Nothing requested (no env vars, no binaries) is inconsistent with that purpose.
Instruction Scope
Instructions are narrowly scoped to producing a structured summary for memory flush/compaction. However the template references tools (exec, read_file, grep_search) and explicit file paths and includes a 'Resume Instruction: do not acknowledge this summary' behavior—both are reasonable for resuming session state but worth noting because they imply the agent may read workspace files and continue work without explicitly informing the user.
Install Mechanism
No install spec and no code files; nothing will be written to disk by an installer. Lowest-risk install posture.
Credentials
The skill references $WORKSPACE and specific project file paths but declares no required environment variables. Access to workspace files is proportionate to the stated task (listing key files), but the missing explicit declaration of WORKSPACE and unspecified file-read behavior means you should confirm what filesystem access the agent will have.
Persistence & Privilege
always:false and default autonomy. The skill does not request permanent presence or privileged modifications to other skills or system-wide settings.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install compact-summary - 安装完成后，直接呼叫该 Skill 的名称或使用
/compact-summary触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v1.0.0
- Introduced compact-summary skill for structured memory compaction in OpenClaw.
- Defines a markdown-based summary format inspired by the <summary> XML from claw-code.
- Captures conversation stats, tools used, user requests, todos, key files, and a timeline.
- Format is applied during memory flush or compaction to improve clarity and retrieval.
- Includes clear instructions for seamless resumption without recapping prior context.
元数据
常见问题
Compact Summary 是什么？
Generates structured, concise memory summaries during flush or compaction, detailing conversation stats, tools, user requests, tasks, files, progress, and ti... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 127 次。
如何安装 Compact Summary？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install compact-summary」即可一键安装，无需额外配置。
Compact Summary 是免费的吗？
是的，Compact Summary 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Compact Summary 支持哪些平台？
Compact Summary 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 Compact Summary？
由 hanxiao-bot（@hanxiao-bot）开发并维护，当前版本 v1.0.0。
推荐 Skills