| name | codex-self-improvement |
|---|---|
| description | Mine recent Codex sessions, memories, rollout summaries, Chronicle if available, existing skills, custom agents, and automations to identify repeated manual workflows worth packaging. Use when the user asks Codex to improve itself, find repeated asks, create or update reusable skills/subagents/automations, or run a Codex-Self-Improvement review. |
Find repeated manual work and convert only the useful, well-evidenced pieces into small reusable assets.
| #!/usr/bin/env python3 | |
| """ | |
| Holo3 TAC UI Eval — test computer-use VLM on TAC surfaces. | |
| Uses dev-browser (https://github.com/SawyerHood/dev-browser) to connect | |
| to your running Chrome with the TAC extension installed, take screenshots | |
| of real extension + SPA surfaces, and send them to Holo3 for evaluation. | |
| Setup: | |
| npm install -g dev-browser && dev-browser install |
| name | openclaw-update |
|---|---|
| description | Safe, deliberate OpenClaw upgrade workflow with pre-flight evaluation, backup, update, and post-verification. Use when the user asks to update or upgrade OpenClaw, says a new release dropped, wants to know if they should update, or asks about version pinning. This skill is the canonical upgrade path: it evaluates whether updating is worth it before touching anything. |
This skill is the full upgrade path — from "should I?" to "done, verified."
Never run openclaw update raw. Always go through this workflow.
A meta-agent with full fleet access: all agents, configs, sessions, and infra. Built around the Context Handoff Protocol (70% threshold).
You are the orchestrator. You have read/write access to all agents, their
configs, sessions, workspaces, and the underlying infra.
Generate any Claude Code command with production-quality patterns, real-world integrations, and intelligent features that make developers feel like they have superpowers.
[command-idea] - Description of the command you want to create[--style] - Generation style:
expert - Like k8s-expert: multiple modes, deep analysis [default]interactive - Heavy user interaction and guided flowsautonomous - Minimal interaction, maximum automation
| find . -type f -printf '%T@ %p\n' | sort -n | tail -1000 | cut -f2- -d" " |
| # Assuming you had a model like this | |
| # | |
| # class Post | |
| # has_attached_file :image, :path => ":rails_root/public/system/:attachment/:id/:style/:filename" | |
| # end | |
| namespace :paperclip do | |
| desc "Recreate attachments and save them to new destination" | |
| task :move_attachments => :environment do |
| # GIT Tricks. | |
| Delete from git all deleted system files: | |
| git rm $(git ls-files --deleted) | |
| View deleted file: | |
| git show HEAD^:path/to/file | |
| git show $(git rev-list --max-count=1 --all -- foo)^:foo |
| --- | |
| - name: Deploy new site release | |
| user: deployer | |
| hosts: all | |
| tasks: | |
| - name: Fetch repo updates | |
| git: > | |
| repo=git@github.com:my/repo.git |
| # Count lines in a files (zsh, bash) | |
| cat big-file.sql | sed '/^\s*$/d' | wc -l | |
| # Count result of search ... | |
| grep 'something' big-file.sql | sed '/^\s*$/d' | wc -l |