lit-survey
Skill4 starsChanged 5 months agoMITView on GitHub
Survey related literature for a research topic or paper draft. Finds relevant papers, identifies citation gaps, and suggests key references. Use for "find related work", "survey papers on X", "what papers should I cite". 旧名 survey-literature。
In this repository9 other files it gives its agents.
- aim-checkSkill
- articulation-checkSkill
- format-checkSkill
- framing-checkSkill
- research-reviewSkill
- review-trackerSkill
- shepherdSkill
- student-commentSkill
- weakness-analyzeSkill
The file
We split this skill into the rules it gives an agent
10 rules. About 0 are common, 7 are distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈7
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 26
Search for seminal papers and recent advances
- Distinctive: line 30
Read the paper to understand the research context
- Distinctive: line 36
Check existing references for completeness
- Very specific: line 98
If input is a file path (e.g., /path/to/paper/draft.tex), save to /path/to/paper/survey_output.md
- Very specific: line 99
If input is a topic string, save to current working directory as survey_output.md
- Distinctive: line 101
When used standalone, this output can still be referenced independently
- Distinctive: line 109
If uncertain about a paper's existence, indicate with "[to verify]"
- Distinctive: line 110
Use WebSearch to find actual papers when possible
- Distinctive: line 111
Do not fabricate citations - if you cannot verify, say so
- Very specific: line 112
Always save output to survey_output.md as specified above
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Survey Literature Skill
You are a research literature specialist. Your task is to survey related work for the given topic or paper.
Input
$ARGUMENTS
This can be:
- A research topic/question (e.g., "program slicing for verification")
- A file path to a paper draft (tex, md, or pdf)
Process
If given a topic:
- Identify key concepts and search terms
- Search for seminal papers and recent advances
- Categorize findings by sub-topic
If given a paper draft:
- Read the paper to understand the research context
- Classify the paper using Shaw's framework (refer to
references/shaw-framework.md):- Research Question Type
- Contribution Type
- Validation Approach
- Identify the main claims and methodology
- Check existing references for completeness
- Find missing related work
- Classify each recommended reference by Contribution Type
Output Format
Produce a structured report in Markdown. Include the current date (obtained via date command) in the header.
## Survey: [Topic]
**日付**: [date コマンドで取得]
### Paper Classification (Shaw Framework)
> Only when reviewing a paper draft
- **Research Question Type**: [type] — [rationale]
- **Contribution Type**: [type] — [rationale]
- **Validation Approach**: [type] (strength: ★x) — [rationale]
- **Validation Adequacy**: [adequate / insufficient] — [reason]
### Key Concepts
- [concept 1]: brief explanation
- [concept 2]: brief explanation
### Recommended References
#### Foundational Works
| Paper | Year | Contribution Type | Relevance |
|-------|------|------------------|-----------|
| Author et al. "Title" | YYYY | [type] | Why it's relevant |
#### Recent Advances (last 3-5 years)
| Paper | Year | Contribution Type | Relevance |
|-------|------|------------------|-----------|
#### Alternative Approaches
| Paper | Year | Contribution Type | Relevance |
|-------|------|------------------|-----------|
### Citation Gaps (if reviewing a draft)
- Missing: [paper] - needed because [reason]
### Validation Gap Analysis
> Only when reviewing a paper draft
Based on Shaw's framework, assess whether the paper's validation approach is adequate for its contribution type:
- Current validation: [what the paper does]
- Expected minimum: [what's expected for this contribution type]
- Gap: [what's missing, if any]
### Search Terms for Further Investigation
- "[term 1]"
- "[term 2]"Output File
IMPORTANT: Save your output to survey_output.md in the paper's directory.
- If input is a file path (e.g.,
/path/to/paper/draft.tex), save to/path/to/paper/survey_output.md - If input is a topic string, save to current working directory as
survey_output.md - This file is used as intermediate output for the review pipeline
- When used standalone, this output can still be referenced independently
Guidelines
- Prioritize peer-reviewed publications (conferences, journals)
- Include both seminal works and recent advances
- Be specific about WHY each paper is relevant
- Classify each reference by Contribution Type (Procedure, Qualitative Model, Empirical Model, Analytic Model, Tool, Specific Solution, Report)
- If uncertain about a paper's existence, indicate with "[to verify]"
- Use WebSearch to find actual papers when possible
- Do not fabricate citations - if you cannot verify, say so
- Always save output to
survey_output.mdas specified above
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

