How do I install the Openclaw Sacred Rules skill?

Run npx skillmds@latest add dvcrn/openclaw-sacred-rules in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Openclaw Sacred Rules safe to use?

This skill has not completed SkillMD's automated safety review yet. Independent scanners report: SkillSpector: PASS, Skill Scanner: PASS. Capability flags: executes scripts. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Openclaw Sacred Rules?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Openclaw Sacred Rules free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Openclaw Sacred Rules?

dvcrn (@dvcrn) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Openclaw Sacred Rules by dvcrn — Agent Skill

Name: Openclaw Sacred Rules
Rating: 4.5 (4 reviews)
Author: dvcrn

Openclaw Sacred Rules

OpenClaw configuration safety rules to prevent system disasters. Use when: (1) Making OpenClaw config changes, (2) Troubleshooting auth failures, (3) Taking system backups, (4) Recovering from config disasters, (5) User asks about config best practices, or (6) Before enabling/disabling system features like sandbox mode.

dvcrn Updated Sep 8, 2026 32 repo stars

File contents

OpenClaw Sacred Rules

The 8 Sacred Rules learned from production disasters. Never violate these rules.

The 8 Sacred Rules

1. Verify Backups Before Confirming

# Always verify files exist and are valid
ls -la backup-file.json
python3 -m json.tool backup-file.json > /dev/null

2. Never Edit openclaw.json Manually

Wrong: nano ~/.openclaw/openclaw.json
Right: Use Python/CLI tools that handle JSON safely

3. Never Add Unverified Config Keys

Check official docs or ask before adding ANY new configuration keys.

4. Never Enable Sandbox Without Backup

Sandbox mode can cascade auth failures. Always backup first and test in isolation.

5. All Providers Failing = Config Issue

If multiple model providers fail simultaneously, suspect recent config changes, not provider outages.

6. Prefix Auth Commands with Environment

source ~/.openclaw/.env && openclaw auth <command>

7. Anthropic 401 = Check auth-profiles.json

Don't read the file directly - use openclaw status or CLI tools.

8. tool_use without tool_result = Reset Session

Corrupted sessions need /reset to recover.

Safe Backup Script

Use the provided backup script instead of manual file copying:

scripts/safe_backup.sh

Cooldown Reset Script

When experiencing "all providers unavailable" errors despite valid credentials, the in-memory cooldown state may be stale. Use:

scripts/reset_cooldowns.sh

This clears expired cooldowns from auth-profiles.json.

Note: This is a workaround for a bug where in-memory cooldown state doesn't refresh when file timestamps expire.

Config Validation

Before applying config changes:

scripts/config_validator.py ~/.openclaw/openclaw.json

Recovery Procedures

When you break a rule, see references/recovery.md for step-by-step recovery procedures.

Auth Troubleshooting

For auth issues, use the safe checker:

scripts/auth_checker.sh

Never directly read auth-profiles.json - use the checker script instead.

---name: openclaw-sacred-rulesdescription: OpenClaw configuration safety rules to prevent system disasters. Use when: (1) Making OpenClaw config changes, (2) Troubleshooting auth failures, (3) Taking system backups, (4) Recovering from config disasters, (5) User asks about config best practices, or (6) Before enabling/disabling system features like sandbox mode.---# OpenClaw Sacred RulesThe 8 Sacred Rules learned from production disasters. **Never violate these rules.**## The 8 Sacred Rules### 1. Verify Backups Before Confirming```bash# Always verify files exist and are validls -la backup-file.jsonpython3 -m json.tool backup-file.json > /dev/null```### 2. Never Edit openclaw.json Manually**Wrong:** `nano ~/.openclaw/openclaw.json` **Right:** Use Python/CLI tools that handle JSON safely### 3. Never Add Unverified Config KeysCheck official docs or ask before adding ANY new configuration keys.### 4. Never Enable Sandbox Without BackupSandbox mode can cascade auth failures. Always backup first and test in isolation.### 5. All Providers Failing = Config IssueIf multiple model providers fail simultaneously, suspect recent config changes, not provider outages.### 6. Prefix Auth Commands with Environment```bashsource ~/.openclaw/.env && openclaw auth <command>```### 7. Anthropic 401 = Check auth-profiles.jsonDon't read the file directly - use `openclaw status` or CLI tools.### 8. tool_use without tool_result = Reset SessionCorrupted sessions need `/reset` to recover.## Safe Backup ScriptUse the provided backup script instead of manual file copying:```bashscripts/safe_backup.sh```## Cooldown Reset ScriptWhen experiencing "all providers unavailable" errors despite valid credentials, the in-memory cooldown state may be stale. Use:```bashscripts/reset_cooldowns.sh```This clears expired cooldowns from auth-profiles.json.**Note:** This is a workaround for a bug where in-memory cooldown state doesn't refresh when file timestamps expire.## Config ValidationBefore applying config changes:```bashscripts/config_validator.py ~/.openclaw/openclaw.json```## Recovery ProceduresWhen you break a rule, see `references/recovery.md` for step-by-step recovery procedures.## Auth TroubleshootingFor auth issues, use the safe checker:```bashscripts/auth_checker.sh```**Never directly read auth-profiles.json** - use the checker script instead.

Frequently asked questions