Memory Defrag

Defragment and reorganize agent memory files: split bloated files, merge duplicates, remove stale information, and restructure the memory hierarchy. Use when...

This skill appears to do exactly what it says — reorganize and prune agent memory files — but it performs destructive actions (deleting or archiving content, removing old tasks, deleting nested directories). Before installing or enabling autonomous use: 1) Review the SKILL.md carefully and decide which destructive rules are acceptable. 2) Require explicit user confirmation for any delete/prune actions (disable autonomous invocation or configure the agent to prompt). 3) Take a backup snapshot of your memory directory (or run the skill in a dry-run mode) before applying changes. 4) Limit periodic automation (cron) until you have tested behavior on test data. 5) If the skill's source/maintainer is unknown, prefer a conservative setup (manual invocation + backups) until you trust it. These steps will reduce the chance of accidental data loss.

Type: OpenClaw Skill Name: memory-defrag Version: 0.1.0 The memory-defrag skill is a utility designed to help an AI agent organize and maintain its internal memory files (e.g., MEMORY.md and the memory/ directory). The instructions in SKILL.md provide a structured process for auditing, splitting, merging, and pruning files to improve efficiency, including safety guidelines such as preserving audit trails and asking for confirmation before destructive changes. There are no indicators of malicious intent, data exfiltration, or unauthorized system access.