openclaw-memory-fusion 安全扫描报告 — 可信 | ClawSafe

5 /100

openclaw-memory-fusion

Build and operate a unified OpenClaw memory system combining native Markdown memory, project memory, structured event memory, and upgrade-safe maintenance workflows

This is a legitimate OpenClaw memory management skill with no malicious behavior detected. It performs only local filesystem operations on designated workspace paths with no shell execution, network access, credential harvesting, or hidden functionality.

技能名称openclaw-memory-fusion

分析耗时35.5s

引擎pi

✓

可以安装

This skill is safe to use. No security concerns identified.

资源类型	声明权限	推断权限	状态	证据
文件系统	`READ,WRITE`	`READ,WRITE`	✓ 一致	Path.read_text/write_text throughout scripts/openclaw_memory_fusion.py
命令执行	`NONE`	`NONE`	—	No subprocess, os.system, or shell execution found
网络访问	`NONE`	`NONE`	—	No curl, wget, HTTP requests, or external connections
环境变量	`READ`	`READ`	✓ 一致	Only reads OPENCLAW_WORKSPACE and OPENCLAW_CONFIG at lines 148-158

目录结构

11 文件 · 97.4 KB · 2524 行

Python 1f · 2320L Markdown 10f · 204L

├─ ▾ 📁 scripts

│ └─ 🐍 openclaw_memory_fusion.py Python 2320L · 91.4 KB

├─ ▾ 📁 templates

│ ├─ 📝 ops_backlog.md Markdown 8L · 121 B

│ ├─ 📝 ops_history.md Markdown 8L · 121 B

│ ├─ 📝 ops_system_projects.md Markdown 7L · 87 B

│ ├─ 📝 project_aliases.md Markdown 8L · 278 B

│ ├─ 📝 project_artifacts.md Markdown 8L · 99 B

│ ├─ 📝 project_decisions.md Markdown 8L · 110 B

│ ├─ 📝 project_overview.md Markdown 16L · 216 B

│ ├─ 📝 project_timeline.md Markdown 8L · 108 B

│ └─ 📝 projects_index.md Markdown 12L · 160 B

└─ 📝 SKILL.md Markdown 121L · 4.7 KB

安全亮点

✓ Pure Python implementation with no shell commands or subprocess calls

✓ Scoped filesystem access to designated ~/.openclaw/workspace directory only

✓ Comprehensive safety defaults with dry-run mode for all mutating operations

✓ Mutation recording with manifests and checkpoints for rollback capability

✓ No network requests, external API calls, or remote code execution

✓ No credential harvesting or sensitive path access

✓ No obfuscation, encoding tricks, or hidden malicious functionality

✓ Template-based output using simple Markdown, no embedded scripts

✓ Clean codebase with no suspicious patterns (base64, eval, atob)