ワンクリックで
cmd-evolve
Cluster MEMORY.md entries by theme and optionally generate SKILL.md drafts.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
メニュー
Cluster MEMORY.md entries by theme and optionally generate SKILL.md drafts.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Scan installed skills to extract cross-cutting principles and distill them into rules — append, revise, or create new rule files.
Audit all installed OpenClaw skills for quality: content overlap, freshness, and uniqueness. Supports Quick Scan (changed only) and Full Stocktake modes.
Export MEMORY.md to a timestamped markdown file for sharing or backup.
Import memory entries from a markdown export file, merging with MEMORY.md without duplicating.
Show current workspace memory: MEMORY.md contents, pending recalls, and index status.
List all OpenClaw agent workspaces with memory and session statistics.
| name | cmd_evolve |
| description | Cluster MEMORY.md entries by theme and optionally generate SKILL.md drafts. |
| user-invocable | true |
| origin | openclaw-mas |
| argument-hint | <project-path> [--generate] |
The first argument is the project path. Before doing anything else:
Reads MEMORY.md, groups related sections by theme, and identifies which clusters
are strong enough candidates to become reusable skills. With --generate, writes
SKILL.md drafts into the project's skills/ directory.
/skill cmd_evolve <project-path> # Analyze and suggest
/skill cmd_evolve <project-path> --generate # Analyze and write SKILL.md drafts
Read ~/.openclaw/workspace-main/MEMORY.md. Parse all ##-level sections —
each section has a heading and a content block.
If MEMORY.md is empty or has fewer than 3 sections, report: "Not enough memory entries to cluster. Add more with /skill cmd_promote." Stop here.
openclaw memory search "pattern workflow behavior"
Include any surfaced recalls alongside MEMORY.md sections in the analysis pool.
Group sections into thematic clusters using:
For each cluster, determine:
## headings)Single-section entries: note them but do not generate from them.
============================================================
EVOLVE ANALYSIS
Source: ~/.openclaw/workspace-main/MEMORY.md
Total sections: <N>
Clusters identified: <K>
============================================================
## CLUSTER 1: "<theme name>"
Sections: <count>
Members:
- <heading 1>
- <heading 2>
Suggested type: <command | skill>
Rationale: <one sentence>
## CLUSTER 2: "<theme name>"
...
## SINGLE-SECTION ENTRIES (not enough to cluster)
- <heading>
## SUMMARY
Command candidates: <N>
Skill candidates: <N>
Single entries: <N>
Suggest next steps:
To generate SKILL.md drafts from these clusters:
/skill cmd_evolve <project-path> --generate
To promote more entries from short-term memory first:
/skill cmd_promote <project-path>
Stop here.
For each cluster with 2 or more sections:
Derive a skill name: lowercase, underscores only, max 32 chars
(e.g. "Adding Database Tables" → add_database_table)
Output path: <project-path>/skills/<skill-name>/SKILL.md
Create the directory if it does not exist.
Write the SKILL.md draft:
---
name: <skill-name>
description: "<one-line description derived from cluster theme>"
user-invocable: <true if command candidate, false if skill>
origin: openclaw-mas
argument-hint: "<project-path>"
evolved_from:
- <MEMORY.md heading 1>
- <MEMORY.md heading 2>
generated: <YYYY-MM-DD>
---
## Project Path
The first argument is the project path. Before doing anything else:
1. Extract the project path from the first argument
2. Verify the path exists
3. Work within that directory for all file operations and shell commands
# <Skill Title>
<Description synthesized from the cluster's memory sections>
## Context
This skill was generated by /skill cmd_evolve from the following MEMORY.md entries:
<list the source sections>
## Workflow
<Step-by-step workflow derived from the clustered memory sections>
## Notes
- Review and refine this draft — it was auto-generated from memory clusters
- Remove the evolved_from and generated frontmatter keys before publishing
After writing each draft:
Generated: <project-path>/skills/<skill-name>/SKILL.md
Final report:
Generated <N> SKILL.md drafts in <project-path>/skills/
Next steps:
- Review each draft and refine the workflow steps
- Remove evolved_from and generated frontmatter keys before deploying
- Install to ~/.openclaw/skills/ when ready