What problem does it solve? Running a multi-stage video pipeline manually requires remembering exact module paths, runtime flags, and project directories, which leads to errors like using the wrong Python interpreter or running from the wrong folder. ## Core Features & Use Cases - Stage-Based Execution: Run the full pipeline or individual stages (ideation, generation, postprocess, upload) with the correct uv run python3 -m src.<stage> command. - Environment Discipline: Enforces uv run python3 over bare python3 so dependencies from pyproject.toml are always resolved, and always changes to the project directory first. - Failure Handling: Surfaces full stderr output, checks for missing .env values or API keys, and supports dry-run isolation of failing stages. - Use Case: You want to regenerate only the postprocess stage of your video pipeline without re-running ideation — invoke the skill with the postprocess stage and it executes the correct command from the right directory. ## Quick Start Run the viral video pipeline's generation stage using the viral-run skill.
