← 返回
AI智能 中文

Jasper ConfigGuard

Safely apply OpenClaw config changes with backup, automatic rollback on failure, health checks, and commands for patching, restoring, listing, diffing, valid...
安全应用OpenClaw配置变更,支持备份、故障自动回滚、健康检查,以及补丁、恢复、列出、对比、验证等命令。
emberdesire
AI智能 clawhub v1.0.0 1 版本 99898.6 Key: 无需
★ 1
Stars
📥 1,951
下载
💾 104
安装
1
版本
#latest

概述

Jasper ConfigGuard v1.0.0

Safe config changes for OpenClaw with automatic rollback. Never brick your gateway again.

Setup

npm install -g jasper-configguard

Usage

Apply a config change safely

jasper-configguard patch '{"gateway":{"bind":"tailnet"}}'

The tool will:

  1. Back up your current config
  2. Apply the patch (deep merge)
  3. Restart the gateway
  4. Wait for health check
  5. Auto-rollback if gateway fails

Preview changes

jasper-configguard patch --dry-run '{"agents":{"defaults":{"model":{"primary":"opus"}}}}'

Restore from backup

jasper-configguard restore

List backups

jasper-configguard list

Check health

jasper-configguard doctor

Agent Integration

Use from your agent to safely modify OpenClaw config:

# Safe model switch
jasper-configguard patch '{"agents":{"defaults":{"model":{"primary":"anthropic/claude-opus-4-5"}}}}'

# Enable a plugin safely
jasper-configguard patch '{"plugins":{"entries":{"my-plugin":{"enabled":true}}}}'

# If something breaks, restore
jasper-configguard restore

API

const { ConfigGuard } = require('jasper-configguard');
const guard = new ConfigGuard();

// Safe patch
const result = await guard.patch({ gateway: { bind: 'tailnet' } });
if (!result.success) console.log('Rolled back:', result.error);

// Dry run
const preview = guard.dryRun({ agents: { defaults: { model: { primary: 'opus' } } } });
console.log(preview.diff);

版本历史

共 1 个版本

  • v1.0.0 当前
    2026-03-28 19:09 安全 安全

安全检测

腾讯云安全 (Keen)

安全,无风险
查看报告

腾讯云安全 (Sanbu)

安全,无风险
查看报告

🔗 相关推荐

ai-intelligence

ontology

oswalpalash
类型化知识图谱,用于结构化智能体记忆与可组合技能。支持创建/查询实体(人员、项目、任务、事件、文档)及关联...
★ 714 📥 244,100
data-analysis

Openclaw Plugin

emberdesire
AI代理推理式入侵检测。结合模式匹配与LLM分析,防御越狱、提示注入、凭证窃取及社会工程。含108种检测模式、OpenClaw插件、自动扫描及隔离功能。命令:hopeid scan/test/setup/stats/doctor。
★ 0 📥 2,897
ai-intelligence

self-improving agent

pskoett
捕获经验教训、错误和纠正,以实现持续改进。使用时机:(1)命令或操作意外失败;(2)用户纠正……
★ 4,061 📥 799,449