skill-autoinstaller

Automatically discover, evaluate, validate, and install new PopeBot skills from GitHub repositories.

Safety Notice

This listing is imported from skills.sh public index metadata. Review upstream SKILL.md and repository scripts before running.

Copy this and send it to your AI assistant to learn

Install skill "skill-autoinstaller" with this command: npx skills add winsorllc/upgraded-carnival/winsorllc-upgraded-carnival-skill-autoinstaller

Skill Auto-Installer

Automatically discover, evaluate, validate, and install new PopeBot skills from GitHub repositories.

Quick Start

Discover skills from a GitHub repo

node /job/.pi/skills/skill-autoinstaller/discover.js --repo zeroclaw-labs/zeroclaw

Evaluate a specific skill

node /job/.pi/skills/skill-autoinstaller/evaluate.js --path /tmp/skills/awesome-skill

Install a skill

node /job/.pi/skills/skill-autoinstaller/install.js --path /tmp/skills/awesome-skill

Full pipeline: discover → evaluate → install

node /job/.pi/skills/skill-autoinstaller/pipeline.js --repo openclaw/openclaw --pattern "skills/*"

Commands

discover.js

Scans a GitHub repository for potential skills.

Scan by repo

node /job/.pi/skills/skill-autoinstaller/discover.js --repo zeroclaw-labs/zeroclaw

Scan local directory

node /job/.pi/skills/skill-autoinstaller/discover.js --dir /path/to/repo

With search pattern

node /job/.pi/skills/skill-autoinstaller/discover.js --repo openclaw/openclaw --pattern "skills/*"

JSON output

node /job/.pi/skills/skill-autoinstaller/discover.js --repo zeroclaw-labs/zeroclaw --json

evaluate.js

Performs security audit and compatibility check on a skill directory.

Evaluate skill

node /job/.pi/skills/skill-autoinstaller/evaluate.js --path /tmp/skills/my-skill

With security audit

node /job/.pi/skills/skill-autoinstaller/evaluate.js --path /tmp/skills/my-skill --security

JSON output

node /job/.pi/skills/skill-autoinstaller/evaluate.js --path /tmp/skills/my-skill --json

install.js

Installs a validated skill into PopeBot.

Install skill

node /job/.pi/skills/skill-autoinstaller/install.js --path /tmp/skills/my-skill

With activation

node /job/.pi/skills/skill-autoinstaller/install.js --path /tmp/skills/my-skill --activate

Dry run

node /job/.pi/skills/skill-autoinstaller/install.js --path /tmp/skills/my-skill --dry-run

pipeline.js

Runs the full discovery → evaluation → installation pipeline.

Full pipeline

node /job/.pi/skills/skill-autoinstaller/pipeline.js --repo zeroclaw-labs/zeroclaw

With pattern matching

node /job/.pi/skills/skill-autoinstaller/pipeline.js --repo openclaw/openclaw --pattern "skills/*"

Interactive mode

node /job/.pi/skills/skill-autoinstaller/pipeline.js --repo zeroclaw-labs/zeroclaw --interactive

Security Audit Checks

The evaluator performs these security checks (inspired by ZeroClaw's audit system):

  • Manifest Validation: SKILL.md must exist with valid frontmatter

  • Path Traversal: No .. in paths or symlinks outside sandbox

  • Dangerous Patterns: No eval, exec, child_process without sandboxing

  • Secret Handling: No hardcoded credentials

  • Network Calls: External calls must be intentional and documented

  • File Operations: Write operations must be confined to /job/tmp/ or skill directory

  • Permission Model: Skills declare required permissions in metadata

Skill Metadata Format

Skills should include this metadata in SKILL.md frontmatter:


name: skill-name description: What the skill does homepage: https://github.com/author/skill metadata: { "popebot": { "emoji": "🔧", "requires": { "bins": ["node", "git"] }, "permissions": ["network", "filesystem"], "install": [ { "id": "npm", "kind": "npm", "package": "@scope/package", "bins": ["tool"], "label": "Install tool (npm)", }, ], }, }

Installation Strategies

The auto-installer supports multiple installation methods:

  • npm: npm install -g package

  • brew: brew install formula

  • apt: apt-get install package

  • pip: pip install package

  • go: go install package@latest

  • binary: Download from release URL

  • script: Run installation script

  • source: Build from source

Error Handling

  • Failed discovery: Returns empty list with error message

  • Failed evaluation: Skill marked as "incompatible" with reasons

  • Failed installation: Rollback to previous state, detailed error log

Output Formats

All commands support --json for machine-readable output:

{ "status": "success", "skills": [ { "name": "skill-name", "path": "/tmp/skills/skill-name", "evaluation": { "passed": true, "score": 95 }, "installed": true } ] }

Source Transparency

This detail page is rendered from real SKILL.md content. Trust labels are metadata-based hints, not a safety guarantee.

Related Skills

Related by shared tags or category signals.

Coding

clipboard-tools

No summary provided by upstream source.

Repository SourceNeeds Review
Coding

clipboard-manager

No summary provided by upstream source.

Repository SourceNeeds Review
Coding

code-review-assistant

No summary provided by upstream source.

Repository SourceNeeds Review
Coding

github

No summary provided by upstream source.

Repository SourceNeeds Review