Andy-HNU/proactive-agent-lite
skills/proactive-agent-lite/SKILL.md
轻量主动代理增强。用于在不显著增加 token 的前提下引入 4 个高价值机制：VBR（先验证再汇报）、Autonomous vs Prompted Cron 区分、工具迁移检查清单、会话压缩恢复流程。适用于避免“改了文本没改机制”、减少自动任务失效、提高会话连续性。
data-ai
Updated Apr 23, 2026
$ install --global
skillsauthnpx skillsauth add Andy-HNU/AndyClaw proactive-agent-liteInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
Security Scan Results
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
3
Scanners Passed
9
Scanners in report
Clean
TrivyContainer and dependency vulnerability scanner
95%
Clean
SemgrepStatic code analysis for vulnerabilities
95%
Clean
mcp-scan (Snyk)Model Context Protocol security validation
95%
Skipped
Snyk (dep)Open source security scanning
50%
Skipped
Socket.devSupply chain security analysis
50%
Skipped
VirusTotalMulti-engine malware detection
50%
Skipped
CrowdStrikeAdvanced threat intelligence
50%
Skipped
OSV-ScannerOpen Source Vulnerability database check
50%
Skipped
OWASP Dep-Check
50%
Last scanned: Apr 23, 2026, 2:13 AM21.7s2 files scanned
SKILL.md
- name:
- proactive-agent-lite
- description:
- 轻量主动代理增强。用于在不显著增加 token 的前提下引入 4 个高价值机制：VBR（先验证再汇报）、Autonomous vs Prompted Cron 区分、工具迁移检查清单、会话压缩恢复流程。适用于避免“改了文本没改机制”、减少自动任务失效、提高会话连续性。
Proactive Agent Lite
只接入高价值、低噪音模块，不启用高频心跳与重主动策略。
1) VBR（Verify Before Reporting）
当准备说“已完成/已修复”前，先执行：
- 验证机制是否真的变更（不是只改文案）
- 验证行为是否符合目标（实际跑通）
- 记录证据（命令输出/文件变更/任务日志）
未验证，不报“完成”。
2) Cron 架构区分
为每个定时任务先判定类型：
- Prompted（提示型）：需要主会话关注/交互
- Autonomous（自治型）：后台脚本可独立完成
默认把“必须发生”的维护任务做成 Autonomous，避免“有提醒没执行”。
3) Tool Migration Checklist
迁移工具/脚本时必须全量检索并更新：
- cron 任务
ops/脚本AGENTS.md/TOOLS.md/HEARTBEAT.md- 相关 skills 与模板
迁移完成后做双验证：
- 旧路径不可用或不再被调用
- 新路径可执行且有日志证据
4) Compaction Recovery（压缩恢复）
触发条件：上下文压缩、会话丢失、用户问“刚聊到哪”。
恢复顺序：
- 读取
memory/YYYY-MM-DD.md（今日/昨日） - 读取
MEMORY.md（主会话） - 读取
memory/archives/*.md（周归档） - 读取
memory/memory.db（主检索源）
恢复后先给简短确认，再继续任务。
5) 不接入项（当前阶段）
- 高频 heartbeat 驱动的主动触发
- “尝试 10 种方法”硬性指标
- 全量自演化/自改写机制
这些功能待周末复盘后再评估。
6) 安全与成本
- 外部机制引入先扫描再接入（semgrep/trufflehog/pip-audit/本地正则）
- 任何主动流程不得绕过用户边界与审批规则
- 日常优先中短回复，避免主动机制造成 token 噪音
References
references/adoption-notes.md
Related Skills
Andy-HNU/telegram-session-compact
tools
VerifiedTrustedCommunity
Compact long Telegram chat sessions and prevent silent non-replies caused by high context usage. Use when a Telegram conversation becomes slow, misses replies, or session tokens are high (for example above ~80-85%), and when the user asks to clean, rotate, or shrink Telegram context.
SKILL.mdUpdated Apr 23, 2026
Andy-HNU/telegram-file-sender
tools
VerifiedTrustedCommunity
Send local files to Telegram chats through OpenClaw CLI delivery. Use when a user asks to send or forward a file (md/txt/pdf/image/video) to Telegram, especially from workspace paths. Supports direct chat targets and group/topic targets.
SKILL.mdUpdated Apr 23, 2026
Andy-HNU/self-improvement-lite
tools
VerifiedTrustedCommunity
轻量自我改进记录技能。用于把错误、用户纠正、知识缺口、功能诉求写入 `.learnings/`，并在每天回顾时将高价值经验提升到 AGENTS.md / SOUL.md / TOOLS.md / MEMORY.md。适用于“避免重复犯错、沉淀经验、降低长期 token 浪费”的场景；默认不启用高频 hooks。
SKILL.mdUpdated Apr 23, 2026
Andy-HNU/policy-news-search
data-ai
VerifiedTrustedCommunity
Search and collect policy/current-affairs source materials before analysis. Use when input is fragmented (keywords, fuzzy title, retell, or URL), when weekly/manual inspection is requested, or when analysis workflow reports critical knowledge gaps. Separates retrieval from analysis and controls agent-browser usage by priority.
SKILL.mdUpdated Apr 23, 2026