Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
Structured workflow for modifying OpenClaw agent core context files including AGENTS.md, SOUL.md, IDENTITY.md, USER.md, TOOLS.md, MEMORY.md, and HEARTBEAT.md. Handles file selection, duplication checks, size limit warnings, format validation, and change verification.
Prevents common config mistakes like placing operational rules in personality files, duplicating content across files, or bloating context files past the truncation limit where they stop loading fully.
When to use it
- Adding a safety rule to block bulk password exports
- Updating agent delegation thresholds
- Refining tone and personality in SOUL.md
- Adding tool-specific conventions to TOOLS.md
- Auditing and trimming an oversized context file near its 20K limit
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Add a rule to never bulk export passwords
- 1Read references/file-map.md to confirm AGENTS.md as the target for safety rules
- 2Check AGENTS.md file size with wc -c to ensure it is under 18K chars
- 3Run grep -i 'password' on AGENTS.md to find existing related rules
- 4Update the existing password manager section instead of adding a duplicate rule
- 5Verify change with grep and report new file size to the user
AGENTS.md updated with bulk export prohibition added to existing password manager rule, new file size confirmed