Super Skill Helper
Paste the following prompt into your AI chat to install this skill:
Install @user_a13d8d43/super-skill-helper according to https://skillhub.cn/install/skillhub.md.
About this skill
Problem
When AI agent skills come from multiple platforms with uneven quality, engineers repeatedly confirm sources, permissions, package structure, and versioning across different CLIs, websites, and repositories. Manual review can miss suspicious commands, overly broad file access, or obfuscated content in SKILL.md, turning skill selection into fragmented work. The goal is to make skill management a reviewable process rather than ad hoc operations.
How it works
Super Skill Helper breaks the skill lifecycle into discovery, creation, vetting, and publishing.
- Discovery: searches candidate skills in a preferred order such as skillhub search, npx skills find, and clawhub search, then flags source trust, SKILL.md content, and external API key requirements.
- Creation: classifies skills into Tool, Workflow, Capability, and Scenario, and keeps SKILL.md concise by moving detailed material into references/.
- Vetting: checks metadata, permission scope, sensitive files, commands, encoded or obfuscated content, and typosquatting. The network + shell combination is treated as high risk.
- Publishing: handles packaging and version management with flows around package_skill.py, and updates the SKILL.md frontmatter.
Boundaries
It fits engineers who filter, audit, and organize AI agent skill packages. If a skill depends on unknown private APIs, complex permission policies, or internal compliance requirements, use it as one checklist and layer in team-level key management, approval flows, and security audits.
Use Cases
- Before adopting a new AI skill, compare candidates on Skillhub, skills.sh, and Cocoloop, then review source trust and permissions.
- When creating a Tool or Workflow skill, scaffold it with `init_skill.py` and move long details into `references/`.
- When auditing a suspicious skill package, check `SKILL.md` for `~/.ssh`, `curl`, Base64 obfuscation, or `network + shell` permissions.
- Before publishing a skill release, package it with `package_skill.py` and update `name`, `version`, and description in `SKILL.md` frontmatter.
Best For
- Platform engineers who evaluate third-party AI skills for a team and need to review source trust, permissions, and suspicious commands.
- Tech leads who turn multi-step agent workflows into `SKILL.md` and need to structure documentation and separate details into `references/`.
- Security engineers who audit skill packages for typosquatting, `shell`/`network` permission requests, or credential-file references.
- Agent engineering leads who package internal skills for publishing and maintain version numbers.
Related Skills
Collects daily conversations, task logs, and errors, then runs a nine-module review to surface blind spots, generate actions, and update memory.
Local long-term memory for AI agents that appends key facts and daily logs, supports cross-session context and search recall, and auto-archives old diaries to gzip files.
Run a grilling session to interact with or test AI agents.
A systematic prompt optimization skill that refines prompts using a four-step distillation framework (diagnose, structure, think, compress) and methodologies from four prompting masters.