sync-skills
Skill24 starsChanged 9 months agoMITView on GitHub
Fetch skills from projects on the current system, genericize them, and add to the claude-skills repository. Use when user says "sync skills", "fetch skills", "import skills from projects", or wants to pull skills from their local projects into this public repo.
23 other projects use a version of this.
Adopted by
Most recently taken up by
- coder013/claude-skillstheir own version, first seen Nov 16, 2026
- morodomi/dev-crewtheir own version, first seen Nov 12, 2026
- devjiwonchoi/skillstheir own version, first seen Nov 8, 2026
- snuffish/dotfilestheir own version, first seen Nov 1, 2026
- dsonders/claude-skillstheir own version, first seen Oct 20, 2026
- SummerSec/SumSec-Skillstheir own version, first seen Oct 15, 2026
Used with
Projects with it also use explain, research and 3 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
sunnyclinches44Skillexplain
Explains code using real-world analogies and ASCII diagrams
Also in 3 projects that use this
NVIDIASkillresearch
Before designing a feature, research how comparable libraries solve the problem, check best practices, known pitfalls, and upstream changes. Use when starting a new feature, investigating an unfamiliar area, or the user asks to research something.
Also in 3 projects that use this
MicrosoftSkillhandoff
Compact the current conversation into a handoff document for another agent to pick up.
Also in 4 projects that use this
SentrySkillcommit
Create commit messages following Sentry conventions. Use when committing code changes, writing commit messages, or formatting git history. Follows conventional commits with Sentry-specific issue references.
Also in 3 projects that use this
GoogleSkillgrill-me
A relentless interview to sharpen a plan or design.
Also in 3 projects that use this
Other versions23 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 23
- Tasihi89/sync-skills★ 29same name
- Terum-Inc/terum-skills★ 28same name
- SummerSec/SumSec-Skills★ 8same name
- fotis89/ctx-launcher★ 4same name
- madnpc/skills★ 3same name
- devjiwonchoi/skills★ 2same name
- paulo-yamagishi/claude-code-setup★ 2same name
- t0ddharris/claude-code-skills★ 2same name
- GalvinGao/skills★ 1same name
- GuanqiHe/claude-skills★ 1same name
13 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository11 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- create-skillSkill
- better-waySkill
- confluence-to-mdSkill
- diagram-generatorSkill
- draft-emailSkill
- jira-ticket-creatorSkill
- md-to-confluenceSkill
- presentation-builderSkill
- teach-meSkill
- youtube-transcriptSkill
The file
Show its rules and text
We split this skill into the rules it gives an agent
13 rules. 0 are common, 9 are distinctive, and 4 are very specific to this project.
0
Common
Most of the 19 edited versions we read keep these.
9
Distinctive
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- could be automatic: a setting could enforce it
- Distinctive: line 20
Search for skills across local projects:
- Very specific: line 26
Default projects root: [your projects root] (e.g.
- Distinctive: line 37
New skill: not present in repo - proceed to genericize
- Distinctive: line 47
If the role is unclear, ask the user before proceeding.
- Distinctive: line 80
No personal/project-specific content remains
- Very specific: line 95
Do not manually edit install.sh - it reads from skills.json.
- Distinctive: line 103
All skills must pass before committing.
- Distinctive: line 107
Add a row to the correct role table in README.md:
- Distinctive: line 121
Commit with a clear message:
- Distinctive: line 129
Don't blindly copy: Always read the source skill first.
- Distinctive: line 131
Don't merge skills: If two projects have different versions of the same skill, ask the user which to keep.
- Very specific: line 135
Don't forget supporting files: Check for references/, assets/, scripts/ subdirectories and copy them too.
- Very specific: line 137
Don't edit install.sh directly: Use init_skill.py to register skills.
“Common” is measured: kept by at least 6 in 10 of the 19 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Sync Skills from Local Projects
Overview
This skill scans projects on the current system for Claude Code skills (.claude/skills/ directories), identifies new or updated skills, genericizes them by stripping project-specific content, and adds them to this repository.
Workflow
Step 1: Identify the Claude Skills Repo
The target repo is the claude-skills project. Confirm its location - it should be the current working directory or specified by the user.
Step 2: Scan for Skills
Search for skills across local projects:
Find all directories matching: <projects-root>/*/.claude/skills/*/SKILL.mdDefault projects root: [your projects root] (e.g. ~/Documents/Projects/). The user may override this with a different path.
For each found skill, collect:
- Skill name (directory name)
- Source project name
- SKILL.md content
- Any files in
references/,assets/,scripts/subdirectories
Step 3: Deduplicate
Compare found skills against skills already in skills/by-role/ and skills/shared/ directories of this repo:
- New skill: not present in repo - proceed to genericize
- Updated skill: present but source has newer/different content - show diff, ask user whether to update
- Identical: skip silently
Step 4: Determine the Role
For each new skill, determine which role folder it belongs in:
Valid roles: pm, engineer, qa, designer, content-creator, devops, shared, leadership, program-delivery-manager, customer-success, recruiter, consultant, sales, marketing, founder, data-scientist, data-engineer, security
If the role is unclear, ask the user before proceeding.
Step 5: Genericize
For each new or updated skill, apply these transformations:
Replace personal references:
- Author names, handles, usernames -> remove or replace with
[your name] - Newsletter/blog names ->
[your newsletter] - Specific domains (e.g.,
name.substack.com) ->[your-domain] - Niche or topic tied to the author ->
[your niche]or[your topic] - Company names ->
[your company] - Team names ->
[your team] - Client names ->
[your client] - Project-specific program names ->
[your program]
Remove project-specific content:
- File paths referencing the source project
- References to other project-specific skills, configs, or workflows
- Environment variables specific to the source project
- API keys, tokens, or credentials (should never exist, but check)
Preserve:
- The skill's core logic, frameworks, and checklists
- Generic examples and templates
- Anti-patterns and quality criteria
- References to general tools or platforms (LinkedIn, Substack, etc.)
- The book or methodology the skill is grounded in
Verify:
- YAML frontmatter has
nameanddescriptionfields - Description has multiple trigger phrases and is pushy
- Body is under 500 lines
- No personal/project-specific content remains
- Skill is self-contained (no broken references to removed content)
Step 6: Add to Repo
For each genericized skill, use init_skill.py to scaffold and register in skills.json:
python3 scripts/init_skill.py <skill-name> --role <role> --description "<one-liner>" --tags "<tag1,tag2>"This creates skills/by-role/<role>/<skill-name>/SKILL.md and adds the entry to skills.json automatically.
Then overwrite the scaffolded SKILL.md with the genericized content.
Do not manually edit install.sh - it reads from skills.json. The registry is the source of truth.
Step 7: Validate
python3 scripts/validate_skill.pyAll skills must pass before committing.
Step 8: Update README
Add a row to the correct role table in README.md:
| `skill-name` | What it does | Book/methodology source |Step 9: Report and Commit
Output a summary:
- Skills added (with source project and role assigned)
- Skills updated (with what changed)
- Skills skipped (already identical)
- Skills needing manual review (couldn't fully genericize)
Commit with a clear message:
git add skills/by-role/<role>/<skill-name>/ skills.json README.md
git commit -m "sync <skill-name> from [source project]"Anti-Patterns
Don't blindly copy. Always read the source skill first. Some skills are deeply project-specific and can't be genericized.
Don't merge skills. If two projects have different versions of the same skill, ask the user which to keep.
Don't strip too much. The goal is to remove personal references, not gut the content. Frameworks, examples, and checklists should survive intact.
Don't forget supporting files. Check for references/, assets/, scripts/ subdirectories and copy them too.
Don't edit install.sh directly. Use init_skill.py to register skills. install.sh reads from skills.json - the hardcoded array no longer exists.
Quality Checklist
Before completing sync:
- Each new skill scaffolded with
init_skill.py(creates folder + registers in skills.json) - Each new skill has valid YAML frontmatter (name + description)
- Description has multiple trigger phrases
- No personal names, domains, or project paths remain
- Placeholder brackets used consistently:
[your niche],[your newsletter],[your-domain],[your name],[your company],[your team],[your client] - README.md skills table updated with new row
python3 scripts/validate_skill.pypasses- Each skill is self-contained and works standalone
- Body under 500 lines per SKILL.md
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.

