agentleFS
Sign inStart free
manastalukdarmanastalukdar/ai-devstudio/CLAUDE.md

ai-devstudio

CLAUDE.md1 starsChanged 7 months agoMITView on GitHub

This software is released under the MIT License. https://opensource.org/licenses/MIT --> AI DevStudio is a comprehensive development environment featuring 203 professional skills for Claude Code CLI that automate repetitive development tasks and save 10-15 hours per week. Key capability areas: Testing & TDD, CI/CD & DevOps, API Development, Database Management, Security & Code Quality, Debugging & Performance, Git Workflows, Documentation, Code Generation, Session Management, MCP Integration. Each skill lives in skills/<name>/SKILL.md with YAML frontmatter. The installer copies all skills to ~/.claude/skills/.…

  • Pipes a download into a shell

The file

We split this CLAUDE.md into the rules it gives an agent

4 rules. About 0 are common, 2 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  • could be automatic: a setting could enforce it
  1. Very specific:

    NEVER use Claude or AI credentials for git commits: All commits must use the developer's own configured git credentials.

    line 67
  2. Distinctive:

    Test skills in real-world projects across different tech stacks

    line 129
  3. Very specific:

    Create detailed skill reference documentation in docs/skills-reference/

    line 130
  4. Distinctive:

    Consider Tier 4+: ML integration, cloud-specific automation, enterprise security

    line 132

≈ “Common” is an estimate: at least 1 in 5 of the 26,980 CLAUDE.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
CLAUDE.md
<!-- Copyright (c) 2025 Manas Talukdar This software is released under the MIT License. https://opensource.org/licenses/MIT -->

AI DevStudio - Professional Claude Code Skills

Project Overview

AI DevStudio is a comprehensive development environment featuring 203 professional skills for Claude Code CLI that automate repetitive development tasks and save 10-15 hours per week.

Key capability areas: Testing & TDD, CI/CD & DevOps, API Development, Database Management, Security & Code Quality, Debugging & Performance, Git Workflows, Documentation, Code Generation, Session Management, MCP Integration.

Architecture

Three-Tier Skill Organization

  • Tier 1 (33 skills): High-impact essentials for immediate productivity
  • Tier 2 (115 skills): Advanced features for professional workflows
  • Tier 3 (17 skills): Power-user tools for specialized capabilities
  • Core (34 skills): Original foundation skills

Skill Structure

Each skill lives in skills/<name>/SKILL.md with YAML frontmatter. The installer copies all skills to ~/.claude/skills/. See .claude/rules/skill-development.md for the full YAML field reference and quality standards.

Project-local utility skills (used by agents/commands, not user-invocable) live in .claude/skills/.

Command → Agent → Skill Architecture

  • Commands (.claude/commands/): Entry points that orchestrate agents and skills
  • Agents (.claude/agents/): Specialized sub-agents with scoped tools and preloaded skills
  • Skills (skills/ + .claude/skills/): Reusable knowledge modules

Agents: code-reviewer, security-auditor, test-runner, claude-md-auditor, quality-fixer

Architecture reference commands (invoke mid-session for full reference):

  • /architecture:skill-system — tiers, YAML fields, naming, token budgets
  • /architecture:hook-events — all 19 hook events, compact reminder hook
  • /architecture:agent-command-system — Command/Agent/Skill hierarchy, agent memory

Persistent Agent Memory

Each agent accumulates institutional knowledge across conversations:

.claude/agent-memory/<agent-name>/MEMORY.md

Loaded into the agent's context on every invocation. Agents append new findings after completion.

Auto-Loaded Rules

Detailed guidelines are in .claude/rules/ (loaded automatically by Claude Code):

FileContents
skill-development.mdYAML frontmatter fields, tier organization, quality checklist
git-workflow.mdCommit rules, credentials policy, branching, checkpoints
code-quality.mdLanguage standards, anti-patterns, install script sync rule
token-optimization.md8 optimization patterns, token budgets, caching strategy

Critical Safety Rules

NEVER use Claude or AI credentials for git commits. All commits must use the developer's own configured git credentials. No AI attribution, no Co-Authored-By: Claude, no git config modifications.

NEVER commit without explicit user request.

Always create a git checkpoint before destructive operations.

Installation

# Mac/Linux
curl -sSL https://raw.githubusercontent.com/manastalukdar/ai-devstudio/main/install.sh | bash

# Cross-platform
python install.py

See README.md for full installation instructions and skill listings.

Current State

  • Skills: 203 implemented
  • Token Optimization: 100% complete — 60-70% average reduction across all skills
  • Tiers: Tier 1 (33), Tier 2 (115), Tier 3 (17), Core (34)

See docs/skills/ for full expansion plan and per-batch implementation summaries.

Usage Patterns

Typical Development Session

claude "/session-start feature-name"
claude "/security-scan"
claude "/scaffold component-name"
claude "/session-update Added OAuth integration"
claude "/review"
claude "/test"
claude "/commit"
claude "/session-end"

Quality Assurance Pipeline

claude "/security-scan" && claude "/review" && claude "/test"
claude "/find-todos" && claude "/fix-todos"

Project Goals

  1. Save 10-15 hours/week through 150 professional skills
  2. Professional-grade code analysis, security scanning, and accessibility compliance
  3. Streamline CI/CD, deployment validation, release automation, and IaC
  4. Enforce TDD workflows, mutation testing, and comprehensive test coverage
  5. Complete API lifecycle: testing → documentation → contract validation
  6. Performance optimization: bundle analysis, profiling, memory leak detection
  7. Enable seamless session continuity across Claude conversations
  8. Universal compatibility across all technology stacks and frameworks
  9. Optimize Claude API costs through intelligent tool usage patterns (60-90% savings)

Next Steps

  • Test skills in real-world projects across different tech stacks
  • Create detailed skill reference documentation in docs/skills-reference/
  • Gather user feedback and usage patterns
  • Consider Tier 4+: ML integration, cloud-specific automation, enterprise security

Development Context

Built upon excellent open-source work — see README.md Acknowledgements for the full list of attributed projects including CCPlugins, claude-sessions, obra/superpowers, Anthropic Skills, MCP, Boris Cherny's multi-agent patterns, and claude-code-best-practice by shanraisshan.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

Reviews can't be read right now.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free