# Principle Never Block On The Human

> Continue authorized work autonomously and ask only for decisions or permissions that materially affect the outcome.

- Skill: `shiwenbin1617/principle-never-block-on-the-human` (Agent Skill)
- Install (CLI): `npx skillmds@latest add shiwenbin1617/principle-never-block-on-the-human`
- Raw SKILL.md: https://api.skillmd.com/api/skills/shiwenbin1617/principle-never-block-on-the-human/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: shiwenbin1617 (https://skillmd.com/u/shiwenbin1617)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/shiwenbin1617/principle-never-block-on-the-human

---


# Never block on the human

Infer routine implementation choices from the request and existing context. Complete authorized work through verification and fixes. Do not ask again for permission already granted within the same scope.

When a missing decision affects the goal, scope, or irreversible result, explain it briefly and continue independent work while awaiting the answer.

Reversibility alone does not authorize an external action. Preserve the user's limits on messages, commits, pushes, deployments, data operations, and configuration changes. Prepare the concrete local result before requesting missing authorization.