j.skillify
Skill0 starsChanged 21 days agoApache-2.0View on GitHub
Refactor one or more existing skills into a cleaner, more maintainable structure — extracting hardcoded content into asset files, offloading deterministic steps into scripts, and reorganizing the skill body to reflect the leaner result. Use this whenever the user wants to clean up, improve, or restructure an existing skill, mentions that a skill has hardcoded content that could be templated, or asks to "skillify", "refactor", or "tidy up" a skill. Trigger even if the user just pastes a skill and says something like "can we make this better?".
In this repository151 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- j.brainstormSkill
- j.breakSkill
- j.btwSkill
- j.check-inSkill
- j.clearifySkill
- j.close-storySkill
- j.cloud-connectSkill
- j.commitSkill
- j.connectSkill
- j.continueSkill
- j.conventionsSkill
- j.dashboard-shareSkill
- j.dashboardSkill
- j.deep-diveSkill
- j.distributeSkill
- j.docSkill
- j.doc-syncSkill
- j.doooSkill
- j.doSkill
- j.jenga-configSkill
- j.jenga-permission-levelSkill
- j.jengaSkill
- j.errorSkill
- j.evaluateSkill
- j.examplifySkill
- j.gitignoreSkill
- j.helpSkill
- j.ideaSkill
- j.improveSkill
- j.initSkill
- j.jbpSkill
- j.lgtmSkill
- j.pi-planSkill
- j.playbook-newSkill
- j.playbookSkill
- j.proceedSkill
- j.publishSkill
- j.reconcile-originSkill
- j.reconcileSkill
- j.redoSkill
- j.smoke-testSkill
- j.spinoffSkill
- j.statusSkill
- j.testSkill
- j.todoSkill
- j.toolsSkill
- j.unchartedSkill
- j.wtfSkill
- skill-nameSkill
- j.brainstormSkill
- j.breakSkill
- j.btwSkill
- j.check-inSkill
- j.clearifySkill
- j.close-storySkill
- j.cloud-connectSkill
- j.commitSkill
- j.connectSkill
- The first 60 shown.
The file
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 4 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 30
Read every SKILL.md the user references.
- Distinctive: line 45
Before making changes, summarise what you found and what you propose to extract.
- Distinctive: line 55
Write the extracted asset files with their content
- Very specific: line 56
Write or update the bash script(s), following the structure in assets/init-new/SKILL.md
- Distinctive: line 62
Keep narrative and conversational steps in the skill body (things Claude decides or explains)
- Distinctive: line 74
Read these when you need a concrete example of the before/after pattern.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Skillify — Refactor a Skill into a Cleaner Structure
skills/j-skillify/ is the canonical, hand-edited directory for this skill, per CLAUDE.md's "The Canonical Naming Contract" (the E50 reopening of 2026-09-09, which promoted skills/j-skillify/ from generated twin to sole canonical form). The j- prefix is there for collision safety — a real directory under a distinct name, so a host tool shipping its own same-named built-in command cannot shadow it (Claude Code's native skill resolution is a literal-string, directory-name-based match; see project/documentation/skill-authoring.md's "Invocation Convention").
⚠️
scripts/generate-j-alias.shwas retired byE50_S14and no longer exists — there is nothing to run. This file was previously generated from a bareskills/skillify/SKILL.mdsource;E50_S15deleted that directory. This file is now the sole canonical, hand-edited source for this skill — edit it directly.
What this skill does
Takes one or more existing skills and refactors them by:
- Extracting hardcoded file content (JSON, Markdown, config, etc.) into
assets/templates - Extracting hardcoded directory lists into
assets/reference files - Moving deterministic multi-step sequences into
scripts/bash scripts - Rewriting the skill body to delegate to those scripts and assets, keeping instructions lean
See assets/init-old/SKILL.md and assets/init-new/SKILL.md for a concrete before/after example.
Instructions
1. Read the skill(s)
Read every SKILL.md the user references. If they also reference scripts or assets already in the skill, read those too.
2. Identify refactor opportunities
For each skill, look for:
| Pattern | Refactor to |
| Hardcoded file content written inline (heredoc, echo, cat) | assets/<name>_template.<ext> copied via cp |
| Hardcoded list of directories or paths | assets/<name>.txt read via a while read loop |
| Sequential deterministic steps (create files, run commands) | scripts/<name>.sh executed by the skill |
SCRIPT_DIR defined repeatedly across steps | Define once at top of script |
3. Confirm the plan with the user
Before making changes, summarise what you found and what you propose to extract. List:
- Which files will move to
assets/ - Which steps will move to
scripts/ - What the updated skill body will look like
Get a thumbs-up before proceeding.
4. Produce the outputs
For each change:
- Write the extracted asset files with their content
- Write or update the bash script(s), following the structure in
assets/init-new/SKILL.md - Rewrite the skill body so each offloaded step becomes a single
cpor script call with a short explanation
5. Update the skill body
The rewritten SKILL.md should:
- Keep narrative and conversational steps in the skill body (things Claude decides or explains)
- Delegate all deterministic file/directory operations to scripts
- Reference asset files by name so it's clear what exists where
- Stay concise — if a step is now just "run the script", say so in one line
Reference files
assets/init-old/SKILL.md— The init skill before refactoring (hardcoded heredocs, inline JSON, hardcoded directory list)assets/init-new/SKILL.md— The init skill after refactoring (all content in assets, steps delegated toinit.sh)
Read these when you need a concrete example of the before/after pattern.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.