daily-agent
长链任务执行skill — 被skill-router调用,负责复杂任务的调度、spawn子代理、Hat系统编排和收尾检查 Skill: daily-agent Owner: paudyyin Summary: 长链任务执行skill — 被skill-router调用,负责复杂任务的调度、spawn子代理、Hat系统编排和收尾检查 Tags: latest:2.18.0 Version history: v2.18.0 | 2026-08-03T00:55:19.437Z | auto daily-agent v2.18.0 - 重大重构:将 daily-agent 定位为被 skill-router 调用的长链任务执行 skill,仅负责复杂任务的调度、子代理 spawn、Hat 系统编排与收尾检查 - 强化主流程为:任务分类 → 复杂度评估 → 执行路由 → 技能匹配 → 委派执行 → 收尾检查 - 场景导航(scene-navigation)升级:基于关键词快速输出匹配场景及对应上下文(减少 token 消耗、提升定位效率) - 前置钩子(P
Rank
62
Safety
84
Downloads
1.2k
Updated
Oct 10, 2026
Version
2.18.0
Source
CLAWHUB
About
What it does, and when to use it.
Capability contract not published. No trust telemetry is available yet. 1.2K downloads reported by the source. Last updated 10/10/2026.
Avoid when
- Contract metadata is missing or unavailable for deterministic execution.
Risk flags: missing_or_unavailable_contract, trust_data_unavailable, schema_references_missing
Public facts
Every fact links back to the source it came from.
- Vendor
- Clawhubvendor · observed Oct 10, 2026
- Protocol compatibility
- OpenClawcompatibility · observed Oct 10, 2026
- Adoption signal
- 1.2K downloadsadoption · observed Oct 10, 2026
- Latest release
- 2.18.0release · observed Aug 3, 2026
- Handshake status
- UNKNOWNsecurity
Install and run
Setup complexity: low.
clawhub skill install s172x1s50dh0bgmm3yc9516nr586kk4b:daily-agent- Setup complexity is classified as HIGH. You must provision dedicated cloud infrastructure or an isolated VM. Do not run this directly on your local workstation.
- Final validation: Expose the agent to a mock request payload inside a sandbox and trace the network egress before allowing access to real customer data.
Contract: missing
curl -s "https://www.xpersona.co/api/v1/agents/clawhub-paudyyin-daily-agent/snapshot"
Documentation
CLAWHUB
144,255 characters of source documentation, loaded on request.
Extracted files
5 files captured from the source.
SKILL.md
---
name: daily-agent
description: "长链任务执行skill — 被skill-router调用,负责复杂任务的调度、spawn子代理、Hat系统编排和收尾检查"
tags: [meta, general, multi-agent, iterative, template-based]
version: 2.18.0
---
# Daily Agent — 长链任务执行skill v2.18.0
**你是长链任务执行skill daily-agent**。由skill-router调用,负责复杂任务的调度和执行。
**定位**:
- 你不是always-on入口,**skill-router才是**
- 你被skill-router调用,专门处理**长链任务**(≥10次调用/复杂调度/Hat系统编排)
- 核心能力:任务分类 → 复杂度评估 → spawn子代理 → Hat系统 → 收尾检查
**核心流程**:任务分类 → 复杂度评估 → 执行路由 → 技能匹配 → 委派执行 → 收尾检查
## 快速决策流程
```
skill-router 调用 → daily-agent 启动
→ Step 0: 模式触发检查(caveman/zoom-out/grill-me/ToT/converse)
→ Step 1.5: 场景导航(Top 3 匹配场景,v2.8 新增)
→ Step 1.8: 前置钩子检查(PreTask hook)(v2.9 新增)
→ Step 2: 复杂度评估(短链/长链 + 工具预算)
→ Step 3: 执行路由(主会话/spawn/cron)
→ Step 4: 技能匹配(基于关键词→技能映射表,daily-agent不直接执行)
→ Step 4.5: 并行执行规划 + 2-Action Rule(v2.10 新增)
→ Step 5: 委派执行(spawn子代理 + 进度反馈)
→ Step 5.5: 重试与降级
→ Step 6: 收尾检查(浏览器/学习/git/记忆/技能审计/规划收尾)
```
---
## Step 0: 内置模式触发检查(v2.5.0 新增)
每条消息进来先检查是否触发内置模式。
| 模式 | 触发条件 |
|------|-----------|
| caveman(压缩沟通) | 用户要求简短/省token/快速回复 |
| zoom-out(全局理解) | 用户要求看全局/整体架构/系统全貌 |
| grill-me(方案审视) | 用户提出方案要求深入追问/审视 |
| ToT(多路径探索) | 用户要求多路径对比/方案探索 |
| **converse(讨论模式)** | 用户想讨论/聊天/探索想法,不要求执行 |
**模式叠加规则**:
1. 模式可叠加,通过 `set_context("mode", mode_name)` 设置
2. 叠加时按最严格的约束执行
3. 退出条件:用户明确说退出或任务完成
**未触发任何模式**:正常进入 Step 0.5/Step 1。
---
## Step 0.5: Converse 模式(v2.5.0 新增)
> 当用户说"讨论一下"或"聊聊"时,他们需要的不是AI执行力,而是对话伙伴。此时禁止任何写操作。
### 意图检测三层判断
**Layer 1: 显式触发词 → 直接进入 CONVERSE 模式**
触发词:`讨论一下` / `聊聊` / `帮我分析` / `你觉得呢` / `聊聊` / `converse` / `怎么样` / `XX怎么样` / `XX行不行` / `帮我想想` / `有看法吗`
→ 自动进入 Converse 模式,输出"进入讨论模式"提示
**Layer 2: 隐式触发句式 → 确认后进入**
触发句式:
- "我在想是不是可以……"(探索性想法,未要求执行)
- "你觉得哪种方案更好?"(征求意见,未要求执行)
- "帮我分析一下XX"(分析请求,未要求执行)
→ 提示:"检测到讨论意图,是否进入 Converse 模式?"
**Layer 3: 上下文推断 → 自动识别**
- 连续 2+ 轮对话中出现探索性/假设性表述且无执行意图
- → 提示:"看起来我们在讨论中,是否继续 Converse 模式?"
### 退出 Converse 模式的条件
用户说出执行意图词 → 退出 CONVERSE 模式,进入执行:
**退出触发词**:`执行` / `开始` / `动手` / `做吧` / `实施` / `go` / `implement` / `开始执行` / `退出 converse`
**退出行为**:立即停止讨论,进入任务调度流程。"执行"、"写代码"、"开始"、"动手"、"做吧"、"实施"等。
### CONVERSE 模式行为约束
1. **禁止写操作**:禁止 write / edit / exec(写操作) / browser(写操作) / message
2. **允许读操作**:read / web_search / web_fetch(仅用于支撑讨论)
3. **优先级**:讨论 > 提供选项 > 分析利弊 > 给出建议
4. **末尾提供推进选项**:
```
---
💬 讨论进行中 | 当前状态:"等待确认" | 下一步
```
5. **可叠加 grill-me/ToT 模式**:讨论中可引入追问或多路径探索
### 状态维护
进入 Converse 模式后维护以下状态:
```json
{
"converse_active": true/false,
"converse_topic": "当前讨论主题",
"discussion_points": ["已讨论的要点"],
"pending_action": "待执行的操作(用户说执行后触发)"
}
```
### 退出时输出格式(v2.5.0 新增)
当用户从讨论切换到执行时,输出 **"讨论→执行"过渡摘要**:
1. **回顾讨论结论**
2. **提取待执行操作**:`pending_action`
3. **输出 CONVERSE 退出提示**:
```
💬 讨论结束。
讨论主题:[topic]
待执行:[pending_action]
进入执行模式。需要我开始吗?输入 **"开始执行"** 启动。
```
### 讨论记录格式(v2.5.0 新增)
Converse 模式结束后生成讨论摘要写入记忆。
```markdown
## Converse 讨_meta.json
{
"ownerId": "kn73z7zhz92tv6skpnmpy5pdfh86j68x",
"slug": "daily-agent",
"version": "2.18.0",
"publishedAt": 1785718519437
}LEARNING.md
# LEARNING.md — Daily Agent 学习记录 ## 使用说明 每次 Paudy 纠正我、指出错误、或我学到新规则时,自动追加到这里。 格式:日期 + 事件 + 学到的教训。 --- ## 学习记录 ### 2026-06-02 — Gateway 重启铁律 - **事件**:反复用 `gateway stop` + 杀进程重启,导致配置被覆盖 - **Paudy 纠正**:应该只用 `gateway restart`(SIGUSR1 热重载) - **教训**: 1. `gateway stop` 连接远程配置服务器 401,经常失败 2. 杀进程再启动会触发 PRD preset 合并,覆盖本地配置 3. SIGUSR1 热重载安全,不杀进程、不合并 preset、配置完整保留 4. `doctor --fix` 自动修复也覆盖配置,禁止自动执行 - **永久规则**:Gateway 重启只用 `gateway restart` ### 2026-06-02 — L1 cursor bug - **事件**:L1 记忆提取从 5/28 起持续产出 0 条 - **根因**:cursor 设为精确 max(recorded_at),下次查询用 `>` 比较排除所有行 - **修复**:cursor +1ms buffer + SELF-CHECK 自动检测 - **教训**:cursor 前进时必须 +1ms 缓冲,避免 `>` 比较排除最新批次 ### 2026-06-02 — 配置防覆盖 - **事件**:memorySearch 和 memory-tencentdb 每次重启后被剥离 - **根因**:PRD preset 里没有这些字段 - **修复**:同步更新 PRD preset + openclaw.json - **教训**:所有自定义配置必须同时写入 PRD preset,否则重启会被覆盖
skill-card.md
## Description: Routes long-running and complex agent tasks by classifying requests, assessing complexity, coordinating subagents, applying Hat-system phases, and performing completion checks. This skill is ready for commercial/non-commercial use. ## Publisher: [paudyyin](https://clawhub.ai/user/paudyyin) ### License/Terms of Use: MIT-0 ## Use Case: Developers and agent operators use this skill to route complex or long-chain tasks, delegate work to subagents, manage planning and review phases, and summarize completion status. ### Deployment Geography for Use: Global ## Known Risks and Mitigations: Risk: The skill broadly orchestrates task routing, local script use, memory/profile writes, and commits, which can create persistent changes without enough user control or scoping. Mitigation: Use only in a tightly controlled workspace after disabling automatic memory/profile writes, automatic commits, unbundled script execution, and always-on activation, or require explicit user approval and diff review for each persistent change. ## Reference(s): - [ClawHub skill page](https://clawhub.ai/paudyyin/skills/daily-agent) - [LEARNING.md](LEARNING.md) - [meeting-minutes.md](templates/meeting-minutes.md) ## Skill Output: **Output Type(s):** [Guidance, Markdown, Code, Shell commands, Configuration] **Output Format:** [Markdown with inline plans, status updates, code blocks, shell commands, and structured summaries] **Output Parameters:** [1D] **Other Properties Related to Output:** [May include delegated task briefs, progress summaries, review results, and generated or modified local files when the host agent permits writes.] ## Skill Version(s): 2.18.0 (source: server release metadata and SKILL.md frontmatter) ## Ethical Considerations: Users should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.
templates/meeting-minutes.md
# 会议纪要模板 ## 基本信息 - **会议名称**: - **日期**: - **时间**: - **地点**: - **参会人员**: - **记录人**: --- ## 会议议程 | 序号 | 议题 | 汇报人 | 时间 | |------|------|--------|------| | 1 | | | | | 2 | | | | | 3 | | | | --- ## 会议内容 ### 议题一:[议题名称] **汇报内容**: **讨论要点**: - **决议/结论**: - ### 议题二:[议题名称] **汇报内容**: **讨论要点**: - **决议/结论**: - --- ## 行动项(Action Items) | 序号 | 行动内容 | 负责人 | 截止日期 | 状态 | |------|---------|--------|---------|------| | 1 | | | | 待开始 | | 2 | | | | 待开始 | | 3 | | | | 待开始 | --- ## 下次会议 - **时间**: - **议题预告**: --- ## 备注
AionUi
Free, local, open-source 24/7 Cowork app and OpenClaw for Gemini CLI, Claude Code, Codex, OpenCode, Qwen Code, Goose CLI, Auggie, and more | 🌟 Star if you like it!
activepieces
AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents
cherry-studio
AI productivity studio with smart chat, autonomous agents, and 300+ assistants.
CopilotKit
The Frontend for Agents & Generative UI. React + Angular
Machine-readable data
The same record, as JSON, for agents and crawlers.
{
"facts": [
{
"factKey": "vendor",
"category": "vendor",
"label": "Vendor",
"value": "Clawhub",
"href": "https://clawhub.ai/paudyyin/skills/daily-agent",
"sourceUrl": "https://clawhub.ai/paudyyin/skills/daily-agent",
"sourceType": "profile",
"confidence": "medium",
"observedAt": "2026-10-10T23:06:48.060Z",
"isPublic": true
},
{
"factKey": "protocols",
"category": "compatibility",
"label": "Protocol compatibility",
"value": "OpenClaw",
"href": "https://www.xpersona.co/api/v1/agents/clawhub-paudyyin-daily-agent/contract",
"sourceUrl": "https://www.xpersona.co/api/v1/agents/clawhub-paudyyin-daily-agent/contract",
"sourceType": "contract",
"confidence": "medium",
"observedAt": "2026-10-10T23:06:48.060Z",
"isPublic": true
},
{
"factKey": "traction",
"category": "adoption",
"label": "Adoption signal",
"value": "1.2K downloads",
"href": "https://clawhub.ai/paudyyin/daily-agent",
"sourceUrl": "https://clawhub.ai/paudyyin/daily-agent",
"sourceType": "profile",
"confidence": "medium",
"observedAt": "2026-10-10T23:06:48.060Z",
"isPublic": true
},
{
"factKey": "latest_release",
"category": "release",
"label": "Latest release",
"value": "2.18.0",
"href": "https://clawhub.ai/paudyyin/daily-agent",
"sourceUrl": "https://clawhub.ai/paudyyin/daily-agent",
"sourceType": "release",
"confidence": "medium",
"observedAt": "2026-08-03T00:55:19.437Z",
"isPublic": true
},
{
"factKey": "handshake_status",
"category": "security",
"label": "Handshake status",
"value": "UNKNOWN",
"href": "https://www.xpersona.co/api/v1/agents/clawhub-paudyyin-daily-agent/trust",
"sourceUrl": "https://www.xpersona.co/api/v1/agents/clawhub-paudyyin-daily-agent/trust",
"sourceType": "trust",
"confidence": "medium",
"observedAt": null,
"isPublic": true
}
],
"events": [
{
"eventType": "release",
"title": "Release 2.18.0",
"description": "## daily-agent v2.18.0 - 重大重构:将 daily-agent 定位为被 skill-router 调用的长链任务执行 skill,仅负责复杂任务的调度、子代理 spawn、Hat 系统编排与收尾检查 - 强化主流程为:任务分类 → 复杂度评估 → 执行路由 → 技能匹配 → 委派执行 → 收尾检查 - 场景导航(scene-navigation)升级:基于关键词快速输出匹配场景及对应上下文(减少 token 消耗、提升定位效率) - 前置钩子(PreTask hook)机制完善,任务分类与场景导航后自动注入 hook 检查,支持自动触发技能/上下文 - 明确角色关系:skill-router 作为入口,daily-agent 仅响应 skill-router 发来的长链/复杂任务 - 源码结构更新:新增 scripts/ 并行与观测脚本,优化 skill 文档,移除过时 skill-card",
"href": "https://clawhub.ai/paudyyin/daily-agent",
"sourceUrl": "https://clawhub.ai/paudyyin/daily-agent",
"sourceType": "release",
"confidence": "medium",
"observedAt": "2026-08-03T00:55:19.437Z",
"isPublic": true
}
]
}Record generated Oct 11, 2026.
