dot-claude

Official

Lock down Claude Code tools with safe hooks.

AuthorCogni-AI-OU
Version1.0.0
Installs0

System Documentation

What problem does it solve?

dot-claude solves the problem of safely configuring Claude Code’s workspace permissions and tool execution behavior so agent-driven edits and command runs don’t hang, over-permission, or perform unintended actions.

Core Features & Use Cases

  • Explicit tool permissions: Define which Bash commands Claude Code is allowed to run using precise glob patterns in .claude/settings.json to prevent blocked or stalled tool use.
  • PreToolUse/PostToolUse safety hooks: Add guardrails and audit logging via PreToolUse and PostToolUse to block dangerous commands (e.g., rm -rf, SQL drops) and trace side effects.
  • Hook-based automation for edits: Trigger follow-up checks (like parsing/edit validation based on file patterns) after commands that modify files.
  • Environment & plugin enablement: Enable required environment variables and language server plugins (e.g., pyright-lsp) to support better tooling during development.

Quick Start

Configure .claude/settings.json by setting narrow Bash permissions, then add a PreToolUse matcher for Bash that blocks destructive command patterns while enabling PostToolUse logging for edit/write actions.

Dependency Matrix

Required Modules

None required

Components

references

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: dot-claude
Download link: https://github.com/Cogni-AI-OU/cogni-ai-agent-skills/archive/main.zip#dot-claude

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.