openclaw-main

OpenClaw main workspace containing agent configuration, long-term memory, identity definition, and tools setup.

Summary

Core workspace for an OpenClaw agent operating as an entrepreneurial partner. Manages agent behavior rules, memory governance, daily operation checklists, and tool configuration. Includes Playwright for web automation tasks.

Included Assets

Core Configuration

• SOUL.md — Agent identity and behavior guidelines
• AGENTS.md — Workspace rules and daily operation checklist
• MEMORY.md — Long-term memory (decisions, preferences, key facts)
• TOOLS.md — Local tool paths and sensitive configuration references

Shared Skills

• Skill library at ~/.openclaw/skills (mapped to cloud sync path)
• Canonical skill source: ai-shared/skills

Dependencies

• playwright (v1.58.2) — Web automation framework

How to Use

This workspace is managed by OpenClaw. The agent reads configuration files silently on session start:

1. SOUL.md — Identity and communication style
2. USER.md — User preferences and habits
3. MEMORY.md — Long-term decisions and facts
4. memory/YYYY-MM-DD.md — Daily operation logs

File Governance Rules:

• MEMORY.md holds only stable rules, indexes, and path references
• Detailed logs go to memory/YYYY-MM-DD.md
• Reports go to memory/reports/
• Topic notes go to memory/<topic>.md

Notes

• Context recovery is internal and silent — agents do not announce reading these files
• All credentials, API keys, and tokens must not be exposed; only paths and status are reportable
• Projects should be placed in /Users/openclaw/workspace/<project>/
• The workspace excludes credentials/ and .env files from backup