# Authority Chain

> Authority Chain

- Skill: `lubochka/authority-chain-3` (Agent Skill)
- Install (CLI): `npx skillmds@latest add lubochka/authority-chain-3`
- Raw SKILL.md: https://api.skillmd.com/api/skills/lubochka/authority-chain-3/raw
- Safety review: pending (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: lubochka (https://skillmd.com/u/lubochka)
- Updated: 2026-09-21
- Page: https://skillmd.com/skills/lubochka/authority-chain-3

---

# Authority Chain

## Purpose
Resolve which instructions win when rules conflict.

## When to Use
Use at the start of every task and whenever instructions disagree.

## Inputs
User request, repository rules, project docs, local skills, and tool constraints.

## Steps
1. Identify all applicable instruction sources.
2. Order them by authority and recency.
3. Name conflicts before acting.
4. Follow the highest-authority applicable rule.

## Gates
No unresolved instruction conflict remains before execution.

## Evidence
The final work reflects the selected rule order.

## Anti-Patterns
Following agent habit over project rules; copying another project's rule file wholesale.

## Related
`project-rules-first`, `document-hierarchy`