Token Diet
>
Notebook Polish
Review and polish Jupyter notebooks (.ipynb) for Data Science work. Use when the user asks to review, clean up or make a notebook presentable — e.g. "revisa mi notebook", "limpia este .ipynb", "deja el notebook presentable", "review my notebook", "clean up this notebook".
Changelog Pro
Generates reliable, hallucination-free changelogs and release notes directly from git history, with Conventional Commits parsing, noise filtering, and a semver suggestion. Use when the user asks for a "changelog", "release notes", or "what has changed since" a tag, commit, or date.
Repo Onboard
Fast, token-cheap orientation report for an unfamiliar repository using a deterministic git-based pipeline (file map, hot files, entry points, activity) with strict read budgets. Use when the user says things like "explícame este repo", "onboarding", "qué hace este proyecto", "explain this repo", or "give me a tour of this codebase".
Eda Quicklook
Fast exploratory data analysis (EDA) of a CSV file using a zero-dependency Python script, followed by an interpretation of data-quality problems and suggested next steps. Use when the user says things like "explore this CSV", "EDA", "what does this data look like", or hands you a CSV and asks for a first impression.