# Survey Literature

> Use when a research question requires comparing multiple papers, mapping a field, or identifying evidence-backed gaps. Triggers on "survey the literature", "文献调研", "find papers about", "what's been done on", "literature review", "related work for".

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

---


# Survey Literature

1. Define the question, inclusion boundary, time range, and comparison axes.
2. Read the local landscape and project findings before searching.
3. Follow `../../references/agent-routing.md` for the `domain-expert` role.
4. Produce a source-backed comparison, synthesis, contradictions, unresolved
   gaps, and project implications. Distinguish source claims from synthesis.
5. Save or update the landscape only when requested.

Coverage claims must name the search boundary. Do not call a convenience sample
systematic or exhaustive.

Use available scholarly integrations, native search/PDF readers and source adapters.
Record per-source coverage (abstract, full text, figure/table or media interval),
publication/retrieval dates and locators. A talk or demo may supplement paper
evidence but cannot replace missing method/results. For close reading, use the
paper packet acquisition/validation contract in
`../read-paper/references/paper-acquisition.md`; retain its schema unchanged.