openclaw-workspace

Summary

A modular OpenClaw workspace for capturing daily memories, knowledge, and personal skills, and for managing multi-agent systems and projects. This workspace centralizes curated knowledge, daily logs, long-term memory, platform tools, and project templates, supporting both personal growth and agent-based automation.

Included Assets

• Knowledge & Memory

  • AGENTS.md — Agent workflow, role instructions, and session routines
  • MEMORY.md — Curated long-term memory (project architecture, decisions, rules)
  • memory/README.md — Memory index and organization guide

• Core Structure

  • SOUL.md — Agent identity and responsibilities
  • TOOLS.md — Environment- or device-specific notes

• Skill, Knowledge, and Tasks

  • archive/knowledge/skill-examples/README.md — Practical skill usage examples
  • archive/tasks/README.md — Multi-agent task execution documentation

• Multi-Agent Platform

  • multi-agent-platform/README.md — Setup and running instructions for the orchestration engine
  • multi-agent-platform/package.json — Node.js dependencies and scripts

• Major Projects (templates and docs)

  • projects/ai-legal-assistant/README.md
  • projects/blockchain-cooperation/README.md
  • projects/blockchain-cooperation/package.json
  • projects/crypto-trading-bot/README.md
  • projects/openclaw-task-dashboard/README.md
  • projects/openclaw-task-dashboard/package.json
  • projects/polymarket-trading-bot/README.md

• Configuration

  • openclaw.json — Default agent and gateway settings
  • package.json — Workspace-level dependencies

How to Use

1. Review Memory and Context

   • Always start a new session by reading SOUL.md, the most recent entries in memory/, and (if in a private human session) MEMORY.md.
   • Use the memory index in memory/README.md to find relevant daily logs or curated summaries.

2. Capture Daily and Long-Term Knowledge

   • Log daily events and important context in memory/YYYY-MM-DD.md.
   • Periodically distill key learnings or decisions into MEMORY.md for future reference.

3. Multi-Agent Operations

   • Use AGENTS.md to guide agent roles, context separation, and workflow routines.
   • Implement or modify agents and task flows following patterns in archive/tasks/README.md.

4. Run the Multi-Agent Platform

   • Enter multi-agent-platform/ and follow setup steps in its README:
     • Install Node.js dependencies and launch orchestrator, worker, scheduler, and dashboard as individual processes.
     • Requires local or remote Redis.
   • Monitor workflows and agent progress via generated markdown dashboards.

5. Project Development

   • Use the guides and templates under projects/ for building AI agents, blockchain systems, trading bots, or dashboards.
   • Each project folder contains its dedicated README.md with goals, task lists, architecture, and setup instructions.

6. Personalize Tools & Skills

   • Record specific devices, SSH hosts, and other environment details in TOOLS.md.
   • Reference or extend documented skills and scripts in archive/knowledge/skill-examples/.

7. Configuration

   • Adjust agent and platform defaults in openclaw.json as needed.

Notes

• Security: Only load MEMORY.md in direct, private user sessions to protect sensitive context.
• Updating Memory: Regularly curate daily logs into long-term memory to prevent bloat and ensure essential details are preserved.
• Project Structure: Each major functional area (tasks, memory, agents, skills, projects) is clearly separated for maintainability and privacy.
• Modularity: You can add or evolve agents, skills, and projects independently within their directories.
• Dependencies: Node.js and some Python skills/scripts are used for project templates and skill examples; check each project's README for details.
• Housekeeping: Clean temporary/duplicate files regularly as advised in memory documentation.

This workspace serves as a robust foundation for personal knowledge management, agent-driven automation, and prototype project launches within the OpenClaw ecosystem.