agentleFS
Sign inStart free
jarosser06jarosser06/drift/CLAUDE.md

drift

CLAUDE.md1 starsChanged 11 months agoMITView on GitHub

TDD framework for AI workflows - define agent/skill/command standards in .drift.yaml, validate project structure programmatically, iterate to compliance. The .claude/ directory contains specialized configurations for AI agents: Agents (.claude/agents/): - cicd.md - CI/CD automation and release management - developer.md - Feature implementation - qa.md - Test writing and coverage Skills (.claude/skills/): - testing - pytest test suite creation (90%+ coverage) - linting - Code quality (black, flake8, isort, mypy) - python-docs - PEP 257 docstrings - code-review - Quality assurance…

  • Reads credentials
  • Installs packages

The file

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

5 rules. About 0 are common, 4 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    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. Distinctive:

    Use pytest fixtures for common test setup

    line 67
  2. Very specific:

    All code quality standards (linting, formatting, type checking) are defined in pyproject.toml.

    line 71
  3. Distinctive:

    Update README.md for user-facing changes

    line 76
  4. Distinctive:

    Keep CLAUDE.md concise (< 1500 tokens)

    line 77
  5. Distinctive:

    Claude Code: Use existing Claude Code CLI (no API key needed)

    line 84

≈ “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

Drift Project

TDD framework for AI workflows - define agent/skill/command standards in .drift.yaml, validate project structure programmatically, iterate to compliance.

Tech Stack

  • Language: Python 3.10+
  • Package Manager: uv
  • Core Dependencies: pydantic, anthropic, boto3, pyyaml, jsonschema
  • Testing: pytest with 90%+ coverage requirement
  • Linting: black, flake8, isort, mypy (see pyproject.toml for configuration)

Project Structure

src/drift/    # Main application code (cli, validation, models)
tests/        # Test suite (unit, integration)
.claude/      # Claude Code configuration (agents, skills, commands)

Claude Code Setup

The .claude/ directory contains specialized configurations for AI agents:

Agents (.claude/agents/):

  • cicd.md - CI/CD automation and release management
  • developer.md - Feature implementation
  • qa.md - Test writing and coverage

Skills (.claude/skills/):

  • testing - pytest test suite creation (90%+ coverage)
  • linting - Code quality (black, flake8, isort, mypy)
  • python-docs - PEP 257 docstrings
  • code-review - Quality assurance
  • github-operations - GitHub automation via MCP

Commands (.claude/commands/):

  • /test - Run pytest tests
  • /lint - Run linters
  • /full-check - Run tests + linting
  • /code-review - Comprehensive code review
  • /create-pr - Create PR with quality validation
  • /create-issue - Create well-structured GitHub issue

Key Commands

# Primary use - TDD workflow (define → validate → fix → iterate)
drift --no-llm           # Fast programmatic validation
drift --format json      # JSON output for CI/CD
drift --rules <names>    # Validate specific rules only
drift list               # List available rules
drift draft --target-rule <rule_name> [--output file.md]   # Generate AI prompt
drift document --rules <rule_name> [--format html]         # Generate documentation
# Development
./test.sh                # Run tests (90%+ coverage)
./lint.sh [--fix]        # Run linters
# Installation
uv pip install -e ".[dev]"

Development Guidelines

Testing Requirements

  • Minimum 90% coverage (enforced by pytest)
  • Comprehensive test coverage for all new features
  • Use pytest fixtures for common test setup
  • Mock external API calls (Anthropic, AWS Bedrock, Claude Code CLI)

Code Quality

All code quality standards (linting, formatting, type checking) are defined in pyproject.toml. Run ./lint.sh before committing.

Documentation

  • PEP 257 compliant docstrings
  • Google-style format with custom parameter separator
  • Update README.md for user-facing changes
  • Keep CLAUDE.md concise (< 1500 tokens)

Provider Support

Drift supports multiple LLM providers:

  • Anthropic API: Set ANTHROPIC_API_KEY
  • AWS Bedrock: Configure with aws configure
  • Claude Code: Use existing Claude Code CLI (no API key needed)

Configuration in .drift.yaml (see README.md for details).

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.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

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