# Compare

> Compare runs, artifacts, methods, models, or claims on a fair and explicit decision basis.

- Skill: `synthetic-sciences/compare` (Agent Skill)
- Install (CLI): `npx skillmds@latest add synthetic-sciences/compare`
- Raw SKILL.md: https://api.skillmd.com/api/skills/synthetic-sciences/compare/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Research & Search
- Author: synthetic-sciences (https://skillmd.com/u/synthetic-sciences)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/synthetic-sciences/compare

---


# Compare

Before comparing, read these two files completely and follow them as the full workflow:

1. `../research-workflows/SKILL.md`
2. `../research-workflows/references/compare.md`

Resolve every target precisely and freeze the comparison criterion before calculating or interpreting differences.