Clawdbot Skill Update

**Quick judgment**: This is a strong, well-scoped operational skill for managing Clawdbot upgrades. It offers a full backup–update–restore pipeline with dynamic workspace detection and good procedural guidance. It is highly suitable for users running Clawdbot in production-like or multi-agent setups who want safer updates and structured rollback options. **Key strengths** - Covers the entire operational flow: check updates → validate → dry-run backup → full backup → update → test → deploy → rollback. - Dynamic discovery of agent workspaces and support for multi-agent, sandboxed environments. - Explicit scripts and documentation (checklists, quick reference) that make repetitive maintenance tasks consistent and less error-prone. **Risks and limitations** - **Data loss risk**: Restore operations overwrite existing config/state; misuse, wrong backup directory, or script bugs could cause loss of recent sessions or configuration changes. - **Credential exposure**: Backups include credentials/auth tokens; improper handling or storage of backup directories can leak secrets. - **Environment-specific**: Tightly coupled to Clawdbot’s directory layout, config format, and pnpm/git workflows; not generally useful beyond that context and may break if those conventions change. **Recommended scenarios** - You frequently update Clawdbot (especially in production or multi-agent setups) and want a repeatable, scripted process with dry-run and rollback. - You need to safeguard complex Clawdbot state (sessions, agents, sandboxes, provider routing) before making configuration or version changes. - You want a quick way to perform emergency restores to the last known-good backup after a failed deployment or a misconfiguration.

**Judgement:** Well-designed, production-oriented skill for managing Clawdbot backups and upgrades, especially in multi-agent setups. It provides a full operational workflow (check, dry run, backup, update, validate, restore) and is suitable for serious Clawdbot operators comfortable with the command line. **Strengths** - Covers all critical data domains: config, sessions, agents, credentials, cron jobs, sandboxes, and git state. - Dynamic workspace detection from the main Clawdbot config enables transparent multi-agent support. - Dry-run backup script plus validation scripts and a step-by-step UPDATE_CHECKLIST reduce the risk of operator mistakes. - Explicit workflows for both standard updates and emergency rollback, including quick-restore from the latest backup. - Documentation is practical and action-oriented, with concrete commands and a clear sequence (check → validate → dry run → backup → stop → update → test → deploy). **Key Risks / Limitations** - **Data overwrite risk during restore:** Running the restore script on the wrong machine or with the wrong backup directory can overwrite existing configs and state; operators must double-check paths and targets. - **Assumed environment:** Assumes a fairly specific layout (e.g., `~/code/clawdbot`, `~/.clawdbot`, `pnpm`, git available), so users with non-standard paths must adapt carefully. - **Secret handling:** Backups include credentials and auth tokens; improper storage or permissions on the backup directory can expose sensitive data. - **CLI skill required:** The workflows are shell- and `pnpm`-driven; less suitable for users uncomfortable with terminal-based maintenance. **Recommended Scenarios** - Production or staging Clawdbot deployments where downtime and data loss are unacceptable, and pre-update backups are mandatory. - Multi-agent or complex setups (multiple providers, sandboxes, routing rules) where manual backup/restore would be error-prone. - Upgrading to versions with breaking changes (e.g., DM lockdown, sandbox scope changes, timestamp format changes) where rollback needs to be guaranteed. - Teams that want a repeatable, documented ops runbook for Clawdbot upgrades and incident recovery.