Index of every page on this site: llms.txt
All skills
acedergren avatar

/publish-skill

@641a79b
Run it yourself

Use when creating a new skill and publishing it to a GitHub repo for installation via npx skills add. Covers scaffold, spec compliance, skill-judge evaluation, commit, push, and install verification. Keywords: publish skill, add skill to repo, npx skills init, share skill, agentic-tools, skills.sh. Triggers on "publish skill" or "share skill".

Use this Skill: https://skilld.dev/gh/acedergren/agentic-tools/publish-skill

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

SKILL.md

≈90 tokens for metadata: the name and description. ≈654 when used: this file.

Description uses 4.5% of example budget

Before choosing a Skill, your Agent reads its name and description. All available Skills share that space.

  • A shorter description leaves more room for other Skills. This entry exceeds our 1% size suggestion.

In our Claude Code example, all Skill names and descriptions share 8,000 characters. This Skill uses ≈361 characters, or 4.5%.

The 1% threshold is a size suggestion. Longer descriptions can still fit.

Your model, settings, and other Skills decide how much text your Agent can read.

Example settings and source

The example uses a 200k-token context and default Claude Code settings. The count includes the name, description, separators, and when_to_use when present. Codex also counts local file paths.

Skit's source and limits: Codex 0.160.1, Claude Code 2.1.292.

Publish Skill

When to Use

Load this skill when the user request matches the frontmatter description for Publish Skill.

Add a new skill to acedergren/agentic-tools and make it installable via npx skills add.

Workflow

1. Scaffold

cd ~/Projects/agentic-tools/skills
npx skills init <skill-name>
# Creates skills/<skill-name>/SKILL.md

2. Write the skill

Edit skills/<skill-name>/SKILL.md. Frontmatter must have exactly two fields:

---
name: skill-name-with-hyphens   # lowercase, letters/numbers/hyphens, ≤64 chars
description: "Use when [trigger conditions]. [WHAT]. Keywords: ..."
---

Description rules — all three required:

  • WHAT: what does it do?
  • WHEN: starts with "Use when..."
  • KEYWORDS: searchable terms at the end

3. Evaluate before committing

Run /skill-judge on the new skill. Target ≥ 84/120 (C grade) before pushing.

Key checks:

  • No "What is X" explanations Claude already knows
  • Has explicit NEVER list with reasons
  • SKILL.md < 300 lines (heavy reference → references/ subdir)
  • No extra frontmatter fields (license, metadata, version)

4. Commit and push

git add skills/<skill-name>/
git commit -m "feat(skills): add <skill-name> skill"
git push origin main

5. Verify installable

npx skills add acedergren/agentic-tools@<skill-name> -g -y

If the skill isn't found, check:

  • Directory name matches name field in frontmatter
  • SKILL.md is at skills/<skill-name>/SKILL.md (not nested deeper)

Install Commands for This Repo

# Single skill
npx skills add acedergren/agentic-tools@<skill-name> -g

# All skills
npx skills add acedergren/agentic-tools --all -g -y

# List available without installing
npx skills add acedergren/agentic-tools -l

NEVER

  • Never push a skill without running /skill-judge first — a skill that fails evaluation won't trigger correctly because its description lacks WHAT/WHEN/KEYWORDS
  • Never use extra frontmatter fields — only name and description are valid; license, metadata, version cause parse errors in some agents
  • Never put "When to use" only in the body — the agent only reads the description to decide whether to load; body content is invisible at trigger time
  • Never name a skill with spaces or special chars — my skill! breaks install; use my-skill

Arguments

$ARGUMENTS: Optional user-provided target, path, environment, symptom, or constraint. When empty, infer the narrowest safe scope from the current repository context and ask only if multiple high-impact choices remain.

Source: SKILL.md on GitHub

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

1 warning7mo4 checks · Risk SAFE
  • Gen Agent Trust Hub8mo

    This skill provides a standard developer workflow for scaffolding, evaluating, and publishing AI agent skills. It utilizes common tools such as Git, npx, and local file management to manage a skill repository owned by the author.

  • Socket8mo

    1 alert: gptSecurity

  • Snyk8mo

    Risk: LOW · No issues

  • ZeroLeaks7mo

    Score: 93/100 · 2 sections analyzed

Signed by skilld at 641a79b. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub last month.

Steadyupdated 6 months ago

README badge

README badge for acedergren/agentic-tools/publish-skill