prisma-systematic-review
Use when conducting a systematic review or structured literature survey — protocol, PRISMA flow, screening log, and synthesis, wired to the project's literature tools. A systematic review is a *method* (predefined protocol, auditable screening, reproducible counts), not an ad-hoc reading pile — for casual exploration or a quick coverage pass, this skill is overkill.
- License 许可
- MIT
- 最近更新
- 2026-09-05
- Stars
- ★ 4
- Forks
- 0
01 前言 Front Matter
- name
- prisma-systematic-review
- license
- MIT
- description
- Use when conducting a systematic review or structured literature survey — protocol, PRISMA flow, screening log, and synthesis, wired to the project's literature tools. A systematic review is a *method* (predefined protocol, auditable screening, reproducible counts), not an ad-hoc reading pile — for casual exploration or a quick coverage pass, this skill is overkill.
02 资源文件 4
- 参考文档references/prisma-2020-checklist.md2.1 KB
- 参考文档templates/prisma-flow.md729 B
- 参考文档templates/protocol-outline.md563 B
- 资产文件templates/screening-log.csv316 B
03 SKILL.md 正文
PRISMA Systematic Review
A systematic review is a method, not a reading pile: protocol first, auditable screening, reproducible counts. Where an ordinary review says "I read some papers and here is my take", a systematic review can answer "how did you search, what did you exclude, and why" — with logs. This skill wires the PRISMA 2020 discipline into the project's literature tools.
Not for: exploring a new area, building a reading list, or drafting an
ordinary Related Work section (writing-related-work when enabled) — those
need no protocol and no flow diagram.
Scaled-down mode: when full PRISMA is heavier than the occasion but coverage still needs to be defensible (e.g. a Related Work section someone may challenge), keep the two minimal records — the queries as executed and the exclusion reasons — and skip the protocol registration and flow diagram. Say that you are doing so.
When to use
- The user wants a systematic review / systematic survey, not an ad-hoc one
- A Related Work section needs defensible coverage ("how did you search?")
- Meta-analysis preparation (screening + data extraction stage)
Files in this skill
references/prisma-2020-checklist.md— the 27-item checklist, condensed; read it when drafting the report, not before.templates/protocol-outline.md— the protocol skeleton (write this FIRST).templates/screening-log.csv— one row per record: source, query, id, decision, reason.templates/prisma-flow.md— the flow-diagram counts file to keep current.
Workflow
- Protocol before searching — draft from
templates/protocol-outline.mdinto the project (research question, PICO/SPIDER framing, inclusion & exclusion criteria, databases, date limits). Persist viaresearch-brief-updateor a protocol file after user confirmation. - Search & log — run
literature-discoverper planned query; record each query and hit count in the screening log. Reproducibility means the queries are written down as executed. - Screen in two passes — title/abstract pass then full-text pass; every
exclusion gets a reason in the log.
literature-readfor abstracts;literature-read-pdf(intensive list) for full texts. - Keep the flow counts current — identified → deduped → screened →
excluded (reasons) → included.
templates/prisma-flow.mdis the file of record; the diagram is generated from it at report time. - Include — papers that pass go to the library only with explicit user
confirmation (
literature-add); otherwise stage them for the session. - Synthesize & report — draft the review following the checklist;
citations via
[@bibkey]/ staged[n];literature-export-bibfor the manuscript.
Done when
- Flow-diagram counts reconcile with the screening log (audited against each other).
- Protocol deviations recorded with dates and reasons.
- Included papers are in the library (user-confirmed) or staged; the bibliography is synced.
- Scaled-down mode: the query log and exclusion reasons exist on disk — the rest is optional.
Rules
- Counts in the flow diagram must reconcile with the screening log — audit them against each other before reporting.
- Deviations from the protocol are allowed but must be recorded with dates and reasons.
- Never silently re-run a query with different terms and keep the old count.
04 同仓库更多 Skills
demo-multi-file
Local test skill with an extra notes file
writing-results
Use when drafting a Results / Experiments / Analysis section — narrating verified evidence from run receipts, with complete statistical reporting, derivation analysis, and honest treatment of negative results. Every number traces to a run; every claim survives the audit.
writing-related-work
Use when drafting a Related Work / literature review section or a standalone synthesis note — organizing prior work into a grounded, cited narrative or comparing methods across papers, from the project library plus staged external discovery. Never cites from memory.
writing-preliminaries
Use when drafting a Preliminaries / Background / Notation section — definitions, problem setup, and formal statements the rest of the paper builds on, with notation consistency and machine-checked formal claims. Reference patterns, not molds; include only what later sections actually use.
writing-methods
Use when drafting a Methods / Methodology / Approach / Model section — expounding the ideas and the reasoning behind the method, grounded in the project's brief, notes, and prior discussions, with empirical details from run receipts. Structure follows the idea's shape; reference angles, not fixed templates.
writing-introduction
Use when drafting or rewriting a paper's Introduction — framing the problem, positioning against prior work, and stating contributions as checkable claims, grounded in the brief, notes, and the library. Reference rhetorical patterns, adapted to the paper — never applied as molds.