Super skill

Skill sdsrss/super-skill/skills/super-skill

Git-backed version control, rollback, and provenance for your Claude Code & Codex Agent Skills — a safety-gated skill package manager (CLI + plugin).

Install

npx -y skills add sdsrss/super-skill --skill super-skill

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

28 days oldThe repository was created 28 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Use when the user wants to manage, version, audit, or roll back their Claude Code Agent Skills, or to review skill candidates mined from past sessions. super-skill is a git-backed package manager for the skills in ~/.claude/skills — it seeds existing skills under version control, explains where each came from, rolls any skill back one command, checks registry integrity (doctor), and surfaces recurring task families worth turning into a skill. Triggers include "version my skills", "roll back that skill", "why do I have this skill", "check my skill registry", "what could become a skill", "mine my sessions".

SKILL.md

2.8 KB, 592 tokens by cl100k_base, as published. Nobody here has run it

super-skill

super-skill is a personal package manager for Agent Skills. It does not run your skills — it manages the ones in ~/.claude/skills: version history, provenance, rollback, integrity checks, and opportunity mining. It is driven by the super-skill CLI (install with uv tool install super-skill-cli or pipx install super-skill-cli; the command stays super-skill).

Scope is deliberately the package-manager form. The self-learning loop (auto-optimizing skills, external distillation) is a deferred research track and is NOT part of this tool — do not imply it exists.

When to use

Version / audit existing skills → super-skill seed (import ~/.claude/skills under version control, read-only on the host) → super-skill list / status.
"Why do I have this skill / where did it come from" → super-skill explain <id> (provenance chain + audit + how to roll back).
A skill regressed, undo it → super-skill rollback <id> [--to vN] (switches the active pointer and re-materializes to ~/.claude/skills).
Registry health → super-skill doctor (content-hash / pointer / host-sync check; read-only). super-skill doctor --fix restores git-recoverable versions and re-materializes drift, then re-verifies.
"What do I keep re-solving that could be a skill" → super-skill mine (task families recurring across ≥3 captured sessions) → super-skill candidate draft → review → super-skill candidate approve <id> (runs two gates, then promotes).
Turn on session capture (feeds mine) → super-skill hooks-config prints the settings.json hooks block for the user to merge (user-global config is theirs to edit). The plugin also ships these hooks in hooks/hooks.json.

How to run it

Run the CLI via Bash and interpret the output for the user. Every write path is explicit and reversible; seed/doctor (no --fix) are read-only. Prefer the plugin's slash commands (/super-skill:status, /super-skill:mine, /super-skill:doctor, /super-skill:candidates, /super-skill:seed) for the common flows.

If super-skill is not on PATH, tell the user to install it: uv tool install super-skill-cli (or pipx install super-skill-cli).

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.