skills.rest
greghavens
greghavensCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

skill-builder

Creates Triggerfish skills with SKILL.md structure and security metadata.

8|1|Updated Feb 9, 2026
Search Tags:#security#customization#AI skills#skill authoring#developer guide#triggerfish
One-click install
npx skills add https://github.com/greghavens/triggerfish --skill skill-builder-greghavens
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: skill-builder
Source: https://github.com/greghavens/triggerfish/tree/main/skills/bundled/skill-builder
Command: npx skills add https://github.com/greghavens/triggerfish --skill skill-builder-greghavens

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill provides comprehensive guidance on creating and managing custom skills for the Triggerfish AI agent platform, enabling users to extend its capabilities securely and efficiently.

Core Features & Use Cases

  • Skill Definition: Learn the SKILL.md format, including mandatory name and description fields, and optional metadata like classification_ceiling, requires_tools, and network_domains.
  • Self-Authoring Workflow: Understand how the agent can create new skills, including the approval process and constraints.
  • Security & Policy: Grasp security scanning mechanisms and how tool/network access is gated by policy.
  • Use Case: You need to create a new skill to monitor a specific API for new data. This Skill teaches you the exact structure, security considerations, and best practices for authoring that new skill.

Quick Start

Use the skill-builder to learn how to create a new Triggerfish skill.

Frequently Asked Questions about skill-builder

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
What is a Triggerfish skill and what structure does it require?▼

A Triggerfish skill is defined by a SKILL.md file containing mandatory name and description fields, alongside optional metadata like classification_ceiling, requires_tools, and network_domains for extending agent capabilities.

Can I restrict network access and tool usage when authoring custom AI skills?▼

Yes, you can gate tool and network access by defining the requires_tools and network_domains fields in the SKILL.md frontmatter, ensuring the custom AI skill adheres to strict security policies.

What security scanning mechanisms are applied during skill authoring?▼

Security scanning mechanisms check your authored skill for prompt injection vulnerabilities and data exfiltration risks, ensuring the custom skill meets the platform's security and policy requirements before approval.

When should I use the classification ceiling field in a SKILL.md file?▼

You should use the classification_ceiling field in your SKILL.md file when a custom skill needs to enforce a strict boundary on the security level or data sensitivity the agent is permitted to handle.

Related Skills

Explore More Software Engineering Skills→
s977043s977043
2

skill-creator

Design a Claude Code skill from a use case and produce a repo-ready package.

Community
Advanced
rongxinzyrongxinzy
21

skill-vetter

Vet AI agent skills for security vulnerabilities and permission scope.

Official
Intermediate
AesopScottAesopScott

tool-policy

Configure allowed-tool sets and execution policies for AI agent safety boundaries.

Community
Advanced
OrcaQubitsOrcaQubits
35

ucp-schema-authoring

Author custom UCP schemas using JSON Schema 2020-12.

Official
Intermediate