Bug-Fix MT Generation Pipeline

Generates "fix bug" type main tasks by introducing mutations into environment codebases, then using integration test failures as task descriptions.

Quick Start

# Generate 5 bug-fix tasks for web_scraping
ct run mtgen-bugfix -e web_scraping --count 5

# With auto-promote to main_tasks/
ct run mtgen-bugfix -e web_scraping --count 5 --promote

The pipeline: ideate (LLM invents a realistic bug) → generate (LLM writes a patch) → validate (Docker: apply patch, run tests, check they fail) → optionally promote (copy to main_tasks/).

Expect ~8% overall validation success rate. Run --count 10 per environment to get 1-2 valid bugs.

How It Works

LLM mode (default): Opus ideates realistic bugs for the environment, generates patches, then validates them in Docker by checking that integration tests break.

CLI Reference

ct run mtgen-bugfix [OPTIONS]

Options:
  -e, --env TEXT       Environment name (required)
  --count INTEGER      Number of tasks to generate (default: 5)
  --promote            Auto-promote valid tasks to main_tasks/
  --model TEXT         Model for LLM pipeline (default: claude-opus-4-6)

Environment Support

The pipeline discovers installed environments with mtgen/bugfix.yml configs.

To add a new environment, you need:

• mutable_paths — where the source code lives
• integration_test_path — test file for the LLM prompt
• docker_compose_path — compose file for validation
• Path transformation in llm_pipeline.py for host-to-container mapping

Validation Output