经 AI Skill Hub 精选评估,开源AI工作流 获评「推荐使用」。这款Agent工作流在功能完整性、社区活跃度和易用性方面表现出色,AI 评分 7.5 分,适合有一定技术背景的用户使用。
Production-tested workspace template for OpenClaw AI agents,提供Memory architecture支持,帮助开发者快速构建AI工作流。
开源AI工作流 是一套完整的 AI Agent 自动化工作流方案。通过可视化的节点编排,将复杂的多步骤任务拆解为清晰的自动化流程,实现全程无人值守的智能处理。支持与数百种外部服务和 API 无缝集成,适合构建数据处理管线、业务自动化和 AI 辅助决策系统。
Production-tested workspace template for OpenClaw AI agents,提供Memory architecture支持,帮助开发者快速构建AI工作流。
开源AI工作流 是一套完整的 AI Agent 自动化工作流方案。通过可视化的节点编排,将复杂的多步骤任务拆解为清晰的自动化流程,实现全程无人值守的智能处理。支持与数百种外部服务和 API 无缝集成,适合构建数据处理管线、业务自动化和 AI 辅助决策系统。
# 方式一:pip 安装(推荐)
pip install openclaw-workspace-template
# 方式二:虚拟环境安装(推荐生产环境)
python -m venv .venv
source .venv/bin/activate # Windows: .venv\Scripts\activate
pip install openclaw-workspace-template
# 方式三:从源码安装(获取最新功能)
git clone https://github.com/kindomLee/openclaw-workspace-template
cd openclaw-workspace-template
pip install -e .
# 验证安装
python -c "import openclaw_workspace_template; print('安装成功')"
# 命令行使用
openclaw-workspace-template --help
# 基本用法
openclaw-workspace-template input_file -o output_file
# Python 代码中调用
import openclaw_workspace_template
# 示例
result = openclaw_workspace_template.process("input")
print(result)
# openclaw-workspace-template 配置文件示例(config.yml) app: name: "openclaw-workspace-template" debug: false log_level: "INFO" # 运行时指定配置文件 openclaw-workspace-template --config config.yml # 或通过环境变量配置 export OPENCLAW_WORKSPACE_TEMPLATE_API_KEY="your-key" export OPENCLAW_WORKSPACE_TEMPLATE_OUTPUT_DIR="./output"
Language: English | 繁體中文
Production-tested workspace template for Claude Code (and OpenClaw) AI agents, extracted from a real agent running 4+ months in production.
A fresh OpenClaw agent is stateless — it wakes up, helps you, and forgets everything. This template turns it into something that persists, learns, and improves over time.
| Capability | Where | What it does |
|---|---|---|
| **Memory journal** | memory/YYYY-MM-DD.md | Daily logs with [hall_*] taxonomy tags for retrieval |
| **Long-term memory** | MEMORY.md | Curated facts, infrastructure, patterns (P0/P1/P2 priority) |
| **AAAK compact** | MEMORY_COMPACT.md | Lossless ~200-token snapshot loaded on every session |
| **Knowledge base** | notes/areas/, notes/resources/ | Topic-organized notes that complement journal entries |
| **Hybrid search** | scripts/memory-search-hybrid.py | BM25 (jieba CJK) × temporal recency × hall-type boost; falls back to keyword overlap |
| **Hall-type tags** | [hall_facts] [hall_events] [hall_discoveries] [hall_preferences] [hall_advice] | Categorize journal entries for boosted retrieval |
| **Self-improvement** | .learnings/, LEARNINGS.md | Track corrections / errors / gaps; auto-promote when recurring ≥ 3 |
| **Memory dreaming** | cron/prompts/memory-dream.md | Weekly cross-domain association of unrelated memories |
| **Memory rumination** | cron/prompts/memory-reflect.md | Weekly contradiction detection with action-tracking + stale-check |
| **Memory expiry** | cron/prompts/memory-expire.md | Monthly auto-archive of memories older than 30 days |
| **Memory janitor** | cron/prompts/memory-janitor.md | Hall-tag backfill + duplicate detection + notes quality check |
| **Cron → flag → hook** | .claude/flags/, .claude/hooks/session-start-flags.sh | Background checks drop flags; next session picks them up |
| **Personality** | SOUL.md, IDENTITY.md, USER.md | Decision priors, name/emoji, user profile |
| **Sub-agent patterns** | guides/sub-agent-patterns.md | Battle-tested delegation with delivery verification |
| **Four defense lines** | AGENTS.md | create → execute → deliver → alert verification chain |
| **Hard-trigger memory search** | .claude/hooks/memory-search-trigger.py | UserPromptSubmit hook forces search on keyword match |
| **Cron system (Mac)** | cron/install-mac.sh | launchd plists with placeholder substitution |
| **Cron system (Linux)** | cron/install-linux.sh | Auto-converts plists → user crontab |
| **Network-wait wrapper** | cron/runner.sh | Waits for network readiness on wake-from-sleep before claude -p |
OpenAI Codex CLI users: Codex is supported as an interactive second-opinion runtime only —cron/is Claude-only by design. See.codex/README.mdbefore runningbootstrap.sh.
1. Install your runtime of choice: ```bash
cron → flag → SessionStart hook for background work triagecurl -fsSL https://openclaw.ai/install.sh | bash
bash scripts/install-cron.sh --install ```
See cron/README.md for details and Post-Install Checklist for verification.
该项目提供了一个开源的AI工作流模板,支持OpenClaw AI agents和Memory architecture,帮助开发者快速构建AI工作流,但需要进一步优化和完善
AI Skill Hub 为第三方内容聚合平台,本页面信息基于公开数据整理,不对工具功能和质量作任何法律背书。
建议在沙箱或测试环境中充分验证后,再部署至生产环境,并做好必要的安全评估。
✅ MIT 协议 — 最宽松的开源协议之一,可自由商用、修改、分发,仅需保留版权声明。
AI Skill Hub 点评:开源AI工作流 的核心功能完整,质量良好。对于自动化工程师和运维人员来说,这是一个值得纳入个人工具库的选择。建议先在非生产环境试用,再逐步推广。
| 原始名称 | openclaw-workspace-template |
| 原始描述 | 开源AI工作流:Production-tested workspace template for OpenClaw AI agents. Memory architecture。⭐8 · Python |
| Topics | workflowai-agentllmmemory-architectureopenclawworkspace-templatepython |
| GitHub | https://github.com/kindomLee/openclaw-workspace-template |
| License | MIT |
| 语言 | Python |
收录时间:2026-05-25 · 更新时间:2026-05-30 · License:MIT · AI Skill Hub 不对第三方内容的准确性作法律背书。
选择 Agent 类型,复制安装指令后粘贴到对应客户端