Skillget

ai-assistant-instructions

agent
Guvenlik Denetimi
Uyari
Health Uyari
  • License — License: Apache-2.0
  • Description — Repository has a description
  • Active repo — Last push 0 days ago
  • Low visibility — Only 6 GitHub stars
Code Gecti
  • Code scan — Scanned 12 files during light audit, no dangerous patterns found
Permissions Gecti
  • Permissions — No dangerous permissions requested
Purpose
This provides a centralized collection of instructions, workflows, and configurations to standardize how AI coding assistants behave across multiple platforms. It acts as a style guide for AI pair programmers.

Security Assessment
The repository poses a very low security risk. The codebase is a collection of markdown files and bash configurations rather than executable software. It does not access sensitive data, contains no hardcoded secrets, and the automated code scan of 12 files found no dangerous patterns. The setup process involves cloning a repository and creating symlinks, which are standard local development actions that do not make external network requests.

Quality Assessment
The project is actively maintained, with its most recent code push occurring today. It is released under the permissive Apache-2.0 license and features CI checks for markdown linting and pre-commit hooks. However, its community visibility is currently quite low. With only 6 GitHub stars, it has not seen widespread community testing or adoption yet. Additionally, the documentation explicitly notes that the project is currently in a transitional migration phase, meaning some legacy files are still being reorganized.

Verdict
Safe to use, though keep in mind the project is actively migrating and has low community visibility.
SUMMARY

A comprehensive, vendor-agnostic framework for consistent AI-assisted development workflows - standardized instructions and commands that work seamlessly across Claude, Gemini, Copilot, and local AI models. See also: https://github.com/JacobPEvans/claude-code-plugins

README.md

AI Assistant Instructions

Teaching AI assistants how to help you better. Yes, it's AI instructions written with AI assistance. We've come full circle.

⚠️ MIGRATION IN PROGRESS: Claude, Gemini, Copilot, and other AI assistant custom commands,
skills, and agents are being moved to JacobPEvans/claude-code-plugins
to be delivered as portable plugins. During this transition, this repository primarily maintains user home
directory configuration files (CLAUDE.md, AGENTS.md, etc.) for generic instructions and standards, and still
includes legacy plugin directories and workflows (for example agentsmd/, /commands, and GitHub workflows)
that will be migrated or removed over time.

License
Markdown Lint
Claude Code Review
pre-commit

What Is This?

A centralized collection of instructions, workflows, and configurations for AI coding assistants.
Drop these into your projects and get consistent, high-quality AI assistance across Claude, Copilot, and Gemini.

Think of it as a style guide, but for your AI pair programmer.

Prerequisites

  • Git 2.30+ (for worktree support)
  • GitHub CLI (gh) 2.0+ (for PR/issue management)
  • (Optional) Python 3.8+ for validation hooks
  • (Optional) Node.js 18+ for markdown linting

Quick Start

# 1. Clone the repo
git clone https://github.com/JacobPEvans/ai-assistant-instructions.git

# 2. Copy agentsmd to your project
cp -r ai-assistant-instructions/agentsmd your-project/

# 3. Create vendor symlinks
cd your-project
ln -s AGENTS.md CLAUDE.md

# 4. Verify setup (if using Claude Code)
claude doctor

Or just browse the documentation and cherry-pick what you need.

Directory Structure

.
├── AGENTS.md                  # Main AgentsMD entry point
├── agentsmd/                  # Supporting files
│   ├── rules/                 # Core principles and standards
│   ├── skills/                # Canonical patterns and rules
│   └── workflows/             # The 5-step development workflow
├── .claude/                   # Claude-specific symlinks
├── .copilot/                  # GitHub Copilot symlinks
├── .gemini/                   # Gemini symlinks
└── .github/                   # GitHub integration (prompts, workflows)

Everything in .claude/, .copilot/, and .gemini/ symlinks back to agentsmd/. One source, multiple consumers. DRY principle in action.

Supported AI Assistants

Assistant Integration Notes
Claude .claude/ directory Full command support via Claude Code
GitHub Copilot .github/copilot-instructions.md + prompts Works in VS Code, GitHub.com, Visual Studio
Gemini .gemini/ directory Style guide and config support

The 5-Step Workflow

This repo centers on a rigorous development workflow:

  1. Research & Explore - Understand before you code
  2. Plan & Document - Write the "what" and "why" before the "how"
  3. Define Success & PR - Set acceptance criteria upfront
  4. Implement & Verify - Build with tests, verify as you go
  5. Finalize & Commit - Clean commits, passing CI

Full details in agentsmd/workflows/.

Key Commands

Initialization & Setup

Command Description
/init-worktree Initialize a clean worktree for development

Git & Repository Management

Command Description
/refresh-repo Merge PRs, sync repo, cleanup stale worktrees
/sync-main Sync current branch with main (use all for all PRs)
/sync-permissions Sync AI assistant permissions to repo

Code Review & Quality

Tool/Skill Description
code-reviewer Task agent Confidence-scored code review with security checks
docs-reviewer Task agent Review and validate documentation

PR Lifecycle Management

Tool/Skill Description
/finalize-pr Complete PR lifecycle management
ci-fixer Task agent Fix CI failures on PRs
superpowers:receiving-code-review Implement code review feedback systematically
/finalize-prs Orchestrate PR finalization across all repositories and report merge-readiness status

Issue & Architecture Management

Tool/Skill Description
/shape-issues Shape raw ideas into actionable GitHub issues
issue-resolver Task agent Analyze and implement GitHub issues
pr-review-toolkit plugin Comprehensive PR review with multi-agent analysis

Delegation & Integration

Command Description
/delegate-to-ai Delegate tasks to external AI assistants

See AGENTS.md for the full list of available skills and tools.

Core Concepts

The documentation covers:

  • Code Standards - Consistency across languages
  • Documentation Standards - AI-friendly markdown
  • Infrastructure Standards - Terraform/Terragrunt patterns
  • Permission System - How AI tool permissions integrate with nix-config
  • DRY Principle - Why everything symlinks to one place
  • Memory Bank - Maintaining AI context across sessions
  • Remote Commit Workflow - Making commits via GitHub API without local clone

Browse agentsmd/rules/ and agentsmd/docs/.

Advanced: This repo integrates with
nix-config for unified permission
management across AI tools. This is optional - the basic setup works
standalone. See agentsmd/docs/permission-system.md
for details.

Need Help?

Contributing

Contributions welcome. See CONTRIBUTING.md for the details, though the short version
is: open a PR, don't be a jerk, and I'll probably merge it.

Security

Found a vulnerability? Please report it responsibly. See SECURITY.md for details.

License

Apache 2.0 - Use it, modify it, just keep the attribution.

Built by a human, refined by AI, used by both.

Yorumlar (0)

Sonuc bulunamadi