Literature reviews
Plan and synthesize a literature review across papers and reports, organizing evidence by research question rather than by source. Use whenever the user asks for a literature review, state-of-the-art overview, evidence map, comparison of studies, research gap analysis, or help structuring sources for an academic or technical review.From its SKILL.md
npx -y skills add shinzoxD/knackbox --skill literature-reviewsAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its file declares
Copied from the file, not written here
The file declares its own license as Apache-2.0. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.
SKILL.md
2.7 KB, 446 tokens by cl100k_base, as published. Nobody here has run it
Literature Reviews
Build a transparent synthesis that explains what the evidence collectively shows, where it conflicts, and what remains unknown.
Workflow
- Define the research question, population or system, outcome, period, and review purpose.
- State inclusion and exclusion criteria before selecting evidence.
- Search across appropriate databases and capture the query, date, and filters.
- Extract comparable fields into an evidence matrix.
- Assess study design, sample, bias, applicability, and confidence.
- Synthesize by theme or question, not one paragraph per paper.
- Identify genuine gaps and distinguish them from topics outside scope.
Evidence matrix
| Source | Design/data | Sample/context | Finding | Limitation | Relevance |
|---|
Output format
# Literature review: <question>
## Scope and method
Question, databases/sources, search date, criteria, and review limitations.
## Evidence summary
Number and types of included sources; overall confidence.
## Findings by theme
For each theme: agreement, disagreement, magnitude, and evidence quality.
## Gaps and open questions
What is not known and what research would resolve it.
## Implications
What the evidence supports doing, avoiding, or testing next.
Rules
- Never claim the search is systematic unless the method and screening process meet that standard.
- Keep study findings separate from author interpretation and your synthesis.
- Preserve effect sizes, uncertainty, and sample context when available.
- Weight evidence by quality and relevance, not citation count.
- Note preprints, retractions, conflicts of interest, and inaccessible full text.
- Do not call a topic a research gap merely because the supplied sources omit it.
Edge cases
- With only a few user-provided sources, call the result a focused synthesis and state that the wider literature was not searched.
- When studies use incompatible outcomes, synthesize narratively rather than forcing a numerical comparison.
- When evidence conflicts, test differences in population, method, definitions, and time period before declaring the field inconsistent.
What ships with it: 1 file
1.6 KB alongside SKILL.md
benchmarks/
- prompts.json1.6 KB