publish-skill
Skill4 starsChanged 3 months agoMITView on GitHub
Package and publish a skill to the skills.sh community registry, a public GitHub repo, or both. Load when the user asks to publish a skill, share a skill publicly, submit a skill to the registry, release a skill, or when universal-skill-creator offers publishing as a final step after creation. Also triggers on "push this skill to skills.sh", "make this skill public", "share this skill", or "contribute this skill to the community". Validates quality before publishing, packages correctly (zip for multi-file, .md for atomic), writes a README if missing, and publishes via npx skills CLI.
- Commits and pushes
13 other projects use a version of this.
Adopted by
Most recently taken up by
- lothlorienblade/agent-skillstheir own version, first seen Nov 8, 2026
- mqmalagris/agent-skillstheir own version, first seen Nov 5, 2026
- djfwilkinson/skillstheir own version, first seen Aug 26, 2026
- Kislonfreak/publish-skilltheir own version, first seen Aug 4, 2026
- skilld-dev/skillstheir own version, first seen Aug 4, 2026
- minilittlechanges-ai/skillstheir own version, first seen Jul 14, 2026
Other versions13 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 13
- glebis/claude-skills★ 414same name
- alexanderop/walkthrough★ 130same name
- drewdrewthis/claude-skills★ 2same name
- skilld-dev/skills★ 2same name
- h-mascot/Enterprise-Crew-skills★ 1same name
- Kislonfreak/publish-skill★ 1same name
- MaxLiuyy/Claude-Code-Skills★ 1same name
- minilittlechanges-ai/skills★ 1same name
- djfwilkinson/skills★ 0same name
- JAVICHILEON/claude---Skills★ 0same name
3 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 repository125 other files it gives its agents.
- AGENTS.mdAGENTS.md
- .cursor/rules/agent-loom-routing.mdcCursor rule
- .cursor/rules/agent-loom-skills-index.mdcCursor rule
- adversarial-hatSkill
- agent-builderSkill
- agent-launcherSkill
- agent-loom-syncSkill
- agent-observabilitySkill
- agent-run-retroSkill
- agent-system-architectureSkill
- api-and-interface-designSkill
- api-deprecation-and-migrationSkill
- apply-paper-to-projectSkill
- app-security-hardeningSkill
- architectural-decision-logSkill
- assumption-mappingSkill
- brainstormingSkill
- browser-testing-with-devtoolsSkill
- business-modelingSkill
- ci-cd-and-automationSkill
- codebase-understandingSkill
- code-review-crspSkill
- code-simplificationSkill
- compress-skillSkill
- context-engineeringSkill
- create-agent-promptSkill
- cross-link-skillsSkill
- customer-discoverySkill
- debug-and-fixSkill
- deep-thinkingSkill
- dependency-mappingSkill
- deploy-anywhereSkill
- deprecate-skillSkill
- design-directionSkill
- design-reviewSkill
- design-systemSkill
- dynamic-routingSkill
- eval-judgeSkill
- eval-outputSkill
- eval-pipelineSkill
- eval-rubric-designSkill
- experimentationSkill
- experiment-backlogSkill
- experiment-readoutSkill
- experiment-runbookSkill
- experiment-specSkill
- fault-localizeSkill
- feature-specSkill
- fermiSkill
- first-principlesSkill
- frontend-designSkill
- generate-changelogSkill
- git-workflow-and-versioningSkill
- gsap-animationSkill
- harness-engineeringSkill
- harness-evolutionSkill
- harness-generationSkill
- idea-evaluationSkill
- idea-generationSkill
- implementation-planSkill
- The first 60 shown.
The file
We split this skill into the rules it gives an agent
10 rules. 0 are common, 6 are distinctive, and 4 are very specific to this project.
0
Common
Most of the 12 edited versions we read keep these.
6
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
- Very specific: line 25
Never publish without a security scan: Before publishing, invoke ALL secure-* skills (discover via ls .agents/skills/secure-*) to scan the …
- Distinctive: line 33
Run validate-skills on the target skill:
- Distinctive: line 38
If it fails → stop. Fix the skill first.
- Distinctive: line 39
If score < 10/14 → stop. Improve the skill first.
- Distinctive: line 40
If the skill contains proprietary content → generalise examples and gotchas before proceeding.
- Distinctive: line 53
Every published skill needs a README.md.
- Very specific: line 107
If the user wants the skill in a public GitHub repo (e.g., dvy1987/agent-loom):
- Very specific: line 129
Check for hardcoded paths (e.g., /home/user/workspace/) in scripts before publishing — they won't work on other machines
- Very specific: line 130
If the skill uses scripts/, verify the script has no hardcoded credentials, internal URLs, or machine-specific paths
- Distinctive: line 195
No changes — citation audit passed.
“Common” is measured: kept by at least 6 in 10 of the 12 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
Publish Skill
You are a skill release engineer. You take a validated, quality-checked skill and publish it to the community — correctly packaged, well-documented, and ready for others to install in one command.
Hard Rules
Never publish without a security scan. Before publishing, invoke ALL secure-* skills (discover via ls .agents/skills/secure-*) to scan the skill. BLOCKED = do not publish. This gate is mandatory and cannot be skipped. Publishing multiplies blast radius — every consumer inherits any vulnerability.
Never publish a skill that fails agentskills validate. Fix the skill first.
Never publish a skill scoring below 10/14. Invoke validate-skills first. If score is below 10, invoke improve-skills before publishing.
Never publish skills containing proprietary context — project-specific URLs, internal API endpoints, company names in examples, or private tool names. These must be generalised before publishing.
Workflow
Step 1 — Pre-publish Validation
Run validate-skills on the target skill:
agentskills validate .agents/skills/<skill-name>/If it fails → stop. Fix the skill first. If score < 10/14 → stop. Improve the skill first. If the skill contains proprietary content → generalise examples and gotchas before proceeding.
Step 2 — Determine Package Format
ls .agents/skills/<skill-name>/- SKILL.md only (Atomic tier) → publish as single
.mdfile - Directory with references/, scripts/, assets/ → package as
.zip
Step 3 — Write or Verify README
Every published skill needs a README.md. If missing, create one:
# [Skill Title]
> One-sentence description of what this skill does.
## Install
npx skills <skill-name> -a codex # Codex
npx skills <skill-name> -a replit # Replit
cp -r <skill-name>/ .agents/skills/ # Universal
## Usage
Trigger: "[example trigger phrase]"
## Platforms
Works with: Codex, Ampcode, Claude Code, Warp, Gemini CLI,
GitHub Copilot, Cursor, VS Code, Replit, Factory.ai, Bolt.new
## License
MITStep 4 — Package
For Atomic skills:
cp .agents/skills/<skill-name>/SKILL.md ./<skill-name>.mdFor Standard/Advanced/System skills:
zip -r <skill-name>.zip .agents/skills/<skill-name>/ \
--exclude "*.pyc" --exclude "__pycache__/*"Step 5 — Publish to skills.sh
npx skills publish ./<skill-name>
# or for zip:
npx skills publish ./<skill-name>.zipConfirm the registry URL returned. Test install:
npx skills <skill-name> -a codex --dry-runStep 6 — Optionally Push to GitHub
If the user wants the skill in a public GitHub repo (e.g., dvy1987/agent-loom):
git add .agents/skills/<skill-name>/
git commit -m "feat: add <skill-name> skill (published to skills.sh)"
git push origin mainStep 7 — Report
Published: <skill-name>
Registry URL: https://skills.sh/<skill-name>
Install command: npx skills <skill-name>
GitHub: https://github.com/dvy1987/agent-loom (if pushed)
Platforms: all agentskills.io-compatibleGotchas
npx skills publishrequires the directory name to match thenamefield in frontmatter exactly — mismatch causes a silent failure- Check for hardcoded paths (e.g.,
/home/user/workspace/) in scripts before publishing — they won't work on other machines - If the skill uses
scripts/, verify the script has no hardcoded credentials, internal URLs, or machine-specific paths
Example
<examples>
<example>
<input>Publish the brainstorming skill to skills.sh</input>
<output>
Running validate-skills on brainstorming...
agentskills validate: ✓
Score: 13/14 — passes 10/14 thresholdChecking for proprietary content... none found.
Package format: single SKILL.md (Atomic tier)
README.md present: yes
Packaging: brainstorming.md
Publishing to skills.sh... npx skills publish ./brainstorming.md
Published: brainstorming Registry URL: https://skills.sh/brainstorming Install: npx skills brainstorming
Test install dry-run: ✓
</output>
</example>
</examples>
Common Rationalizations
| Excuse | Reality |
| "Internal validate is enough" | Publish blast radius needs full security sweep. |
| "Redact later" | Secrets in published skills are permanent incidents. |
| "Community fork is fine" | Publish gate exists because consumers trust our namespace. |
Verification
- Full
secure-*family returned SAFE agentskills validate+ line count ≤200 on published skill- No API keys, tokens, or private paths in output
- Changelog or release note prepared
Reference Files
references/publish-checklist.md: Full pre-publish checklist including proprietary content scan, platform compatibility check, and registry submission requirements. Read for any System-tier skill before publishing.
Red Flags
- Directory name mismatches frontmatter name field
- Hardcoded local paths left in scripts or examples
- Publish attempted before full secure-* library sweep
- Scripts contain internal URLs or machine-specific paths
Prune Log
Last pruned: 2026-07-04
- No changes — citation audit passed; content current (improve-skills full pass 2026-07-04)
Impact Report
Published: [skill-name] Registry URL: https://skills.sh/[skill-name] Install command: npx skills [skill-name] Package format: [.md / .zip] validate-skills score at publish: [N]/14 Proprietary content scan: clean GitHu...
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.

