domains / research / llm-judge-evaluation / advanced-evaluation
The two copies are near-identical in structure, taxonomy, bias list, metric table, and prompt templates. The muratcankoylan copy is the more complete and operational version: it adds explicit activation boundaries, a deterministic pre-check step before pairwise judging, and a tie-breaker rationale requirement. The sickn33 copy is a shortened/mirrored variant that omits those steps and adds only a citation and a risk tag. Crowning the fuller copy does not hide a real alternative procedure.
| Repository | Path |
|---|---|
| muratcankoylan/Agent-Skills-for-Context-Engineering | skills/advanced-evaluation/SKILL.md |
| Role | Repository | Path |
|---|---|---|
| inferior | sickn33/agentic-awesome-skills | plugins/agentic-awesome-skills-claude/skills/advanced-evaluation/SKILL.md |
| adapter | sickn33/agentic-awesome-skills | plugins/agentic-awesome-skills/skills/advanced-evaluation/SKILL.md |
| adapter | sickn33/agentic-awesome-skills | plugins/agentic-bundle-aas-ai-product-evaluation-ops/skills/advanced-evaluation/SKILL.md |
| adapter | sickn33/agentic-awesome-skills | skills/advanced-evaluation/SKILL.md |