Proactive Agent
让 AI 助手从「等待任务」变成「主动预见」的架构设计。包含 WAL 协议、Working Buffer、自动定时任务等经过实战验证的模式。
Install
npx skills add wangyendt/wayne-skills --skill proactive-agentgit clone --depth 1 https://github.com/wangyendt/wayne-skills.git /tmp/wayne-skills
mkdir -p ~/.claude/skills
cp -r /tmp/wayne-skills/proactive-agent ~/.claude/skills/proactive-agentPersonal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.
From the SKILL.md
# 主动式 Agent 🦞 **By Hal Labs** — Hal Stack 的一部分 **一个主动预见需求、持续自我进化的 AI 助手架构。** 大多数 AI 只是等待指令。这个 AI 会预判你的需求,并且随着时间推移越来越懂你。 ## v3.1.0 更新 - **自动任务 vs 提醒任务** — 搞懂什么时候用 `systemEvent`,什么时候用 `isolated agentTurn` - **验证实现,而非意图** — 检查机制是否真的变了,别只看文字 - **工具迁移检查清单** — 废弃工具时,确保所有引用都更新 ## v3.0.0 更新 - **WAL 协议** — 写前日志:捕获修正、决策和重要细节 - **Working Buffer** — 在上下文压缩的危险区存活 - **压缩恢复** — 上下文被截断后如何逐步恢复 - **统一搜索** — 说「我不知道」之前,搜遍所有来源 - **安全加固** — 技能安装审核、Agent 网络警告、上下文泄露防范 - **坚持不懈的足智多谋** — 试 10 种方法后再求助 - **自我改进护栏** — 用 ADL/VFM 协议安全进化 --- ## 三大支柱 **主动 — 创造未被请求的价值** ✅ **预见需求** — 问「什么能帮到我的主人?」而不是干等 ✅ **反向提示** — 抛出你没想到但可能感兴趣的想法 ✅ **主动检查** — 关注重要事项,需要时主动提醒 **持久 — 扛过上下文丢失** ✅ **WAL 协议** — 关键细节先写下来,再回复 ✅ **Working Buffer** — 危险区的每条消息都记录 ✅ **压缩恢复** — 上下文被压缩后知道怎么恢复 **自进化的 Agent 进化** ✅ **自我修复** — 先修自己的问题,才能帮你 ✅ **足智多谋** — 放弃前试 10 种方法 ✅ **安全进化** — 护栏防止跑偏和复杂度膨胀 --- ## 目录 1. [快速开始](#快速开始) 2. [核心哲学](#核心哲学) 3. [架构概览](#架构概览) 4. [记忆架构](#记忆架构) 5. [WAL 协议](#wal-协议) ⭐ 新 6. [Working Buffer 协议](#working-buffer-协议) ⭐ 新 7. [压缩恢复](#压缩恢复) ⭐ 新 8. [安全加固](#安全加固) 9. [坚持不懈的足智多谋](#坚持不懈的足智多谋) 10. [自我改进护栏](#自我改进护栏) 11. [自动任务 vs 提醒任务](#自动任务-vs-提醒任务) ⭐ 新 12. [验证实现,而非意图](#验证实现而非意图) ⭐ 新 13. [工具迁移检查清单](#工具迁移检查清单) ⭐ 新 14. [六大支柱](#六大支柱) 15. [心跳系统](#心跳系统) 16. [反向提示](#反向提示) 17. [成长循环](#成长循环) --- ## 快速开始 1. 复制资源文件到工作区:`cp assets/*.md ./` 2. 你的 Agent 检测到 `ONBOARDING.md`,会主动了解你 3. 回答问题(一次性或断断续续都可以) 4. Agent 自动填充 USER.md 和 SOUL.md 5. 运行安全审计:`./scripts/security-audit.sh` --- ## 核心哲学
Details
- Skill name
proactive-agent- Repository
- wangyendt/wayne-skills
- Path
proactive-agent/SKILL.md- License
- MIT
- Repo updated
- 23 October 2026
FAQ
What does the Proactive Agent skill do?
让 AI 助手从「等待任务」变成「主动预见」的架构设计。包含 WAL 协议、Working Buffer、自动定时任务等经过实战验证的模式。
How do I install the Proactive Agent skill?
skills CLI (Claude Code, Cursor, Codex and more): npx skills add wangyendt/wayne-skills --skill proactive-agent. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.
Is Proactive Agent popular?
It lives in wangyendt/wayne-skills, which has 9 GitHub stars and 0 forks.
Related skills
Alapi
9wangyendt/wayne-skills
Comprehensive ALAPI skill covering the full bundled ALAPI OpenAPI catalog. Use when the user mentions ALAPI, alapi, ALAPI_TOKEN, openapi.json, or an ALAPI path like /api/..., or asks to find which ALAPI API matches a business need, choose between similar ALAPI endpoints, explain request fields, prepare or execute a request, or troubleshoot an ALAPI response. Covers inspirational content, articles, news and hotlists, AI chat/translation/NLP, OCR and document recognition, weather and calendar data, IP/phone/ID/domain/ICP/enterprise lookup and verification, finance/exchange/gold/oil/crypto market
Android App Publisher
9wangyendt/wayne-skills
Build or publish standalone Android APKs to an authenticated OSS distribution service, update the latest release, list published apps, and generate stable download links and QR codes. Use for Android app sharing and release management, not generic OSS operations or app-store uploads.
Awesome Docs
9wangyendt/wayne-skills
Use this skill whenever the user asks to record, organize, update, or maintain project documentation, including architecture overviews and decisions, technical books and derivations, API/protocol/configuration references, implementation plans, roadmaps, todos, risks, experiments, know-how, or common commands. Choose paths and naming by document purpose, preserve existing canonical locations, and check links when reorganizing. Defaults to the target project’s docs directory unless the user specifies another location. Covers reader-first technical explanations and interactive HTML documents as w
Product Metrics
9tarunccet/pm-skills
Define a complete product metrics framework: North Star metric selection, input metrics, dashboard design, alert thresholds, and — for AI products — a four-layer AI metrics stack (model quality, operational, product-level, business). Use when setting up metrics for a new product, choosing a North Star, designing a dashboard, or defining KPIs for an AI feature.
Monetization Models
9tarunccet/pm-skills
Brainstorm 3-5 monetization strategies with audience fit, risks, and validation experiments. Use when exploring revenue models, evaluating pricing strategies, or deciding how to monetize a product.
Prompt Engineering
9tarunccet/pm-skills
Craft, evaluate, and manage production-quality prompts including system prompts, few-shot examples, chain-of-thought instructions, and guardrails. Use when designing prompts for a product feature, improving output quality, reducing hallucinations, or building a prompt versioning strategy.
Browse the most popular Claude skills, all ai, llms & agents skills, or write your own with the skills generator.