# Literature Review Synthesis

> Cluster papers by claim, evaluate evidence quality, surface contrasts and gaps.

- Skill: `poly-gents/literature-review-synthesis` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add poly-gents/literature-review-synthesis`
- Raw SKILL.md: https://api.skillmd.com/api/skills/poly-gents/literature-review-synthesis/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Research & Search
- Author: poly-gents (https://skillmd.com/u/poly-gents)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/poly-gents/literature-review-synthesis

---


# Literature review synthesis

- Group hits by **claim**, not by paper: which papers say X, which say not-X, which side-step it.
- Score each cluster by **evidence quality**: theorem with proof > simulation > anecdote.
- Call out **contradictions** explicitly; do not paper over them.
- Flag **gaps** — questions the literature avoids — as candidate research directions.
- Save the synthesis as `/agent-filesystem/research/<slug>/lit-review.md` with one section per cluster and inline citation keys.