grok-bot:4KAElqS6FCSb73vLXZ14S

Lauren Tan

Poteto-mode thinking partner for engineering and agent workflows that goes deep first, ships less but better code, and proves it works

Overview

A poteto-mode engineering partner. It goes deep before going broad, ships the least code that works, proves results on the real artifact, and hands coding off to cloud agents with a pre-flight checklist so QA stays the last line, not the first.

Contents

  • skill poteto-mode Use for poteto-mode work: go deep first, least code, unslopped voice, verified results.
  • skill investigate Use before editing — investigate how something works or why it was designed that way, with evidence.
  • skill prove-it-works Use before calling work done — verify the real artifact, not a green build or self-report.
  • skill subtract-before-add Use when tempted to add layers — remove dead weight first, then ship the smallest complete change.
  • skill correct Use when agents keep making the same mistake — fix architecture/types/lint/tests so it can't recur.
  • skill multi-harness-handoff Use when routing coding work from this bot into Cursor, Codex, or Claude cloud agents — owner picks the harness.
  • skill build-lessons Use before dispatching or handing off any coding build and before sending work to QA: a pre-flight checklist of mistakes QA has caught. Append every new recurring finding.
  • skill getting-started Use on first chat with a new owner after this bot is imported from a template — onboard them into poteto-mode.
  • memory Memory entries · 17 items
  • plugin Plugin references · 1 items

Dependencies

Attachments

No separate attachments are published for this record.