AgriciDaniel/skill-forge / skills/skill-forge-publish
skill-forge-publish
VerifiedPackage and distribute Claude Code skills for sharing via GitHub, Claude.ai uploads, or team deployment. Creates install scripts, documentation, and .skill packages. Use when user says "publish skill", "share skill", "package skill", "distribute skill", or "release skill".
Skill Publishing & Distribution
Process
Step 1: Pre-Publish Validation
Run the full review before publishing:
- Execute
/skill-forge review <path>and ensure score >= 80/100 - Fix any critical or high-priority issues
- Test with at least 5 trigger queries
- Verify all cross-references resolve
Step 2: Create Install Script
Generate install.sh that handles:
#!/usr/bin/env bash
# Install script for [skill-name]
# Usage: bash install.sh
set -euo pipefail
SKILL_DIR="$HOME/.claude/skills"
AGENT_DIR="$HOME/.claude/agents"
SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
echo "Installing [skill-name] skill..."
# Create directories
mkdir -p "$SKILL_DIR" "$AGENT_DIR"
# Copy main skill
cp -r "$SCRIPT_DIR/skill-name" "$SKILL_DIR/"
echo " Installed: skill-name"
# Copy sub-skills
for skill in "$SCRIPT_DIR/skills"/skill-name-*/; do
if [ -d "$skill" ]; then
skill_basename=$(basename "$skill")
cp -r "$skill" "$SKILL_DIR/"
echo " Installed: $skill_basename"
fi
done
# Copy agents (if any)
if [ -d "$SCRIPT_DIR/agents" ]; then
cp "$SCRIPT_DIR/agents"/*.md "$AGENT_DIR/" 2>/dev/null || true
echo " Installed agents"
fi
echo ""
echo "Installation complete!"
echo "Test with: /skill-name"
Step 3: Create README.md (repo-level, NOT inside skill folder)
# [Skill Name]
[1-2 sentence description focusing on outcomes, not features]
## What it does
[Bullet list of key capabilities]
## Installation
### Claude Code
git clone https://github.com/[user]/[repo] cd [repo] bash install.sh
### Claude.ai
1. Download the latest release (.zip)
2. Go to Settings > Capabilities > Skills
3. Click "Upload skill"
4. Select the downloaded .zip file
## Commands
| Command | Description |
|---------|-------------|
| `/skill-name` | [description] |
| `/skill-name cmd` | [description] |
## Examples
### [Example 1 title]
User: "[example input]"
[Description of what happens and expected output]
## Architecture
[file tree diagram]
## License
[License type]
Step 4: Package for Distribution
For Claude.ai upload:
Run python scripts/package_skill.py <path> <output-dir> to create a .skill zip file.
For GitHub:
- Create repository with README.md at root
- Skill folder(s) at root level
- install.sh at root level
- Add LICENSE file
- Add .gitignore (exclude .tmp/, pycache/, *.pyc)
For team deployment (Claude.ai admin):
- Skills can be deployed workspace-wide by admins
- Package as .skill zip and upload through admin console
Step 5: Create .gitignore
__pycache__/
*.pyc
*.pyo
.tmp/
*.egg-info/
dist/
build/
.env
*.skill
Step 6: Release Checklist
- All files validated (score >= 80)
- install.sh tested on clean system
- README.md covers installation, usage, and examples
- LICENSE file included
- .gitignore configured
- No secrets or API keys in any file
- Test queries documented
- Version tagged (if using git)
Step 7: Post-Publish
After publishing:
- Test installation from scratch on a clean environment
- Run all trigger test queries
- Collect initial user feedback
- Plan first iteration based on feedback
- Set up issue templates for bug reports
Distribution Channels
| Channel | Best For | Format |
|---|---|---|
| GitHub | Open source, community | Repository + install.sh |
| Claude.ai upload | Personal use | .skill zip |
| Team admin | Organization-wide | .skill zip via admin console |
| Claude Plugin Marketplace | Wide distribution | .claude-plugin/ manifest |
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| writing-skillsobra/superpowersUse when creating new skills, editing existing skills, or verifying skills work before deployment | Stars: 319k | Updated | Claude CodeCodex | Verified | Install |
| xlsxanthropics/skillsUse this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .xltx, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like "the xlsx in my downloads") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |
| shipping-and-launchaddyosmani/agent-skillsPrepares production launches. Use when preparing to deploy to production, or when asking what needs to be in place before shipping. Use when you need a pre-launch checklist, when setting up monitoring, when planning a staged rollout, or when you need a rollback strategy. | Stars: 118k | Updated | Claude CodeCodex | Verified | Install |
| childhoodthedotmack/claude-memBabysit a shipped release after merge, publication, or deployment. Verify published packages, production behavior, installed plugins, and late CI or user reports until the observation window is clean; use babysit for PR readiness before release. | Stars: 118k | Updated | Claude CodeCodex | Verified | Install |