# Principle Human Owns The Ends

> Reserves goals and shipping decisions for the user. Apply when deciding whether autonomous work may proceed or finish.

- Skill: `bostonaholic/principle-human-owns-the-ends` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add bostonaholic/principle-human-owns-the-ends`
- Raw SKILL.md: https://api.skillmd.com/api/skills/bostonaholic/principle-human-owns-the-ends/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: bostonaholic (https://skillmd.com/u/bostonaholic)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/bostonaholic/principle-human-owns-the-ends

---


# The Human Owns the Ends

Keep two decisions human: what to build and what to ship; run everything between autonomously, with PR review as the checkpoint.

- Never pause an autonomous run to triage findings.
- Loop Blocking and Major findings through the fixer automatically.
- Put Minor-and-below findings in PR review notes; Minor is the human's queue, not a wastebasket.
- Never land on the system's own judgment; open PRs autonomously, but require a human merge decision.
- Defer questions that can wait to PR review.
- Halt blocked runs terminally and report instead of asking permission to continue.