config-checkpoint_skill

This skill provides git-based checkpoints and rollback for OpenClaw configurations, safeguarding changes before sensitive operations.
  • Python

2.5k

GitHub Stars

2

Bundled Files

2 months ago

Catalog Refreshed

3 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill openclaw/skills --skill config-checkpoint

  • _meta.json290 B
  • SKILL.md11.9 KB

Overview

This skill provides git-based version control for OpenClaw configuration changes, enabling safe checkpoints and reversible restores. It creates descriptive commits before sensitive operations and offers staged rollback options, with mandatory checks for .gitignore and sensitive files.

How this skill works

The skill inspects the OpenClaw configuration repository (~/.openclaw/) and runs pre-commit checks: verify .gitignore, scan for sensitive files, and preview staged changes. For SAVE it stages (targeted or full) and commits with a descriptive message; for ROLLBACK it lists recent commits and performs soft, mixed, or hard reset based on user confirmation, warning explicitly before destructive actions.

When to use it

  • Before modifying core files (SOUL.md, AGENTS.md, openclaw.json, etc.)
  • Before installing or uninstalling skills in workspace/skills/
  • Before making bulk memory or cron changes
  • When the user requests an undo of recent configuration changes
  • When recovering system state after failures or corruption

Best practices

  • Always verify ~/.openclaw/.gitignore exists before SAVE
  • Scan untracked/modified files for .pem, .key, .token, credentials/ or secret strings before committing
  • Use targeted git add paths instead of git add -A to avoid accidental commits
  • Prefer git reset --soft or --mixed before considering --hard
  • Require explicit, single-word confirmation (e.g., "yes") before any hard reset

Example use cases

  • Create a checkpoint before installing a new skill: save workspace/skills/ and commit with a message
  • Undo a recent skill installation by listing commits and performing a soft reset to preserve staged files
  • Recover from a broken configuration by rolling back to a known-good commit after reviewing diffs
  • Automated health checks read git status and alert if uncommitted changes exist (read-only)
  • Batch memory edits: SAVE first, apply changes, and ROLLBACK if tests fail

FAQ

The skill stops the SAVE operation and asks you to create or verify ~/.openclaw/.gitignore to prevent committing sensitive files.

Can autonomous triggers perform hard resets?

No. Hard resets are never allowed autonomously; ROLLBACKs always require explicit user confirmation.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational