Integrity Agent Tools
Integrity’s Agent Tools repository is a shared collection of Claude Code skills, security hooks, and output styles maintained by the dev team. It provides a curated toolkit that makes working with Claude Code safer, faster, and more consistent across projects.
Why Agent Tools?
The toolset addresses three problems that come up when a team adopts agentic AI coding:
- Safety — Pre-built security hooks catch common mistakes (leaked secrets, unsafe commands) before they cause damage.
- Speed — Skills automate repetitive workflows so you spend less time on scaffolding and more time on the problem.
- Consistency — Shared output styles and conventions keep AI-generated code aligned with Integrity’s standards, no matter who is prompting.
What’s Included
The repository organizes its resources into three categories:
- Skills — Slash commands (
/skill-name) that extend Claude Code with project-specific workflows and multi-step automations. Run them directly in your Claude Code session. Think of them as reusable playbooks for common tasks. - Hooks — Claude Code hooks that run automatically at key points (e.g., before a commit or after a tool call) to enforce safety checks and team policies.
- Output Styles — Formatting presets that control how Claude Code presents its responses, keeping output readable and consistent.
Each category has its own directory in the repo with documentation alongside the tools themselves.
Getting Started
Browse the repository to see what’s available. Each tool includes its own documentation with setup instructions — copy the ones that are useful to you into your own Claude Code configuration. The repo’s README has full details on every skill, hook, and style.
Interested in contributing? See the Contributing Guide for how to add or update tools.