Building an SLR with Claude Code

This is a systematic literature review project. Create the following directory structure:

- 01_search_and_screening/
- 02_data_extraction/
- 03_synthesis_and_writing/
- 04_manuscript_drafts/
- 00_literature_files/

Then populate CLAUDE.md with this SLR protocol:

# Systematic Literature Review Protocol

## Core Task
Systematic review to understand [YOUR RESEARCH QUESTION]

## File Structure
- /01_search_and_screening/ - Search strings, screening logs
- /02_data_extraction/ - Extraction tables (CSV/Markdown)
- /03_synthesis_and_writing/ - Theme analysis, drafts
- /04_manuscript_drafts/ - Final paper sections
- /00_literature_files/ - Full-text papers (markdown)

## Inclusion Criteria
- [YOUR CRITERIA - e.g., Peer-reviewed, 2020-2025, empirical]
- [Geographic/industry scope]
- [Methodology requirements]

## Exclusion Criteria
- [YOUR CRITERIA - e.g., Non-English, opinion pieces, duplicates]

## Citation Style
APA 7th Edition

## Custom Commands
- /screen: Evaluate abstract against inclusion criteria
- /extract: Extract data from full-text paper
- /critique: Review writing for quality and style

smith2024a.md
jones2024b.md
lee2023.md

"Analyze @00_literature_files/smith2024a.md"
"Compare @00_literature_files/smith2024a.md and @00_literature_files/jones2024b.md"

"Apply the /screen command to every abstract in
@01_search_and_screening/database_export_abstracts.txt
and output results to 01_search_and_screening/screening_results.csv"

"Read the exported Markdown @04_manuscript_drafts/full_draft.md
and the verified Zotero/Better BibTeX bibliography. Resolve each citation
key against that bibliography; list missing or ambiguous keys instead
of inventing entries. Format the verified entries in APA 7th edition."

"Search all files in @02_data_extraction/ and count
occurrences of: 'scalability', 'scaling', 'growth'"

"Run /extract on @00_literature_files/smith2024a.md
and append results to @02_data_extraction/master_extraction_table.md"

"Using data in @02_data_extraction/master_extraction_table.md,
draft a 'Emerging Themes' section. Synthesize findings across
studies and identify patterns."

"Reconcile the search, record, screening, retrieval and study-mapping
ledgers in @01_search_and_screening/ before drafting a PRISMA flowchart.
Keep records, reports and studies distinct. Stop on missing evidence."

"Create a todo list for synthesizing these 5 papers on organizational learning:
- Identify common themes
- Extract methodological approaches
- Map theoretical frameworks
- Identify research gaps
- Draft synthesis section"

"Based on all papers in @02_data_extraction/master_extraction_table.md,
what are the three most significant theoretical gaps? For each gap,
explain why it's important."

"My main argument is [YOUR ARGUMENT]. Act as a skeptical reviewer
from [TARGET JOURNAL]. Formulate three powerful critiques based on
the evidence I've synthesized."

"Based on limitations across all papers, propose a research agenda
with three novel questions that address these gaps. Justify why each
represents a meaningful contribution."

Evaluate this abstract against the inclusion criteria in @CLAUDE.md.

Input: $ARGUMENTS
Include the stable record ID, title, authors, year, and abstract. If an ID is missing, ask for it rather than inventing one.

## Screening Decision

Decision: [INCLUDE | EXCLUDE | UNCERTAIN]

Reason: [Specific criterion met/failed]

Confidence: [High | Medium | Low]

Notes: [Any boundary cases or uncertainties]

Return one quoted CSV row using these columns:
ID, Title, Authors, Year, Decision, Reason, Confidence
Preserve the input ID. Use UNCERTAIN when the abstract does not settle a criterion; send that record to human review. Escape quotes within a field by doubling them.

/screen ID: S001; Title: Organizational scaling; Authors: not supplied; Year: 2023; Abstract: This paper examines organizational scaling in tech startups
using survey data from 500 companies across Europe and Asia between
2020-2023...

Extract structured data from this paper:

Paper: $ARGUMENTS

## Extraction Template

| Field | Content |
|-------|---------|
| Stable record ID | Preserve the screening ID |
| Report ID and citation key | Use the reviewed report-to-study mapping and Zotero key |
| Source filename | Exact supplied Markdown file |
| Authors & Year | |
| Research Question | |
| Methodology | [Qual/Quant/Mixed] |
| Sample | [n=?, context] |
| Data Collection | |
| Analysis Approach | |
| Key Findings | [3-5 bullets] |
| Theoretical Contribution | |
| Limitations | |
| Future Research | |

For each substantive field, include its page/section locator in the supplied source. Write "not reported" when the source does not settle it; do not infer missing sample or findings. Extract only after a human has confirmed full-text inclusion; UNCERTAIN is not a final inclusion.

Append this table to @02_data_extraction/master_extraction_table.md

/extract @00_literature_files/smith2024a.md

Review this writing section for quality:

Section: $ARGUMENTS

## Quality Checklist

1. Logical Flow: Are arguments well-structured?
2. Evidence Strength: Are claims supported by citations?
3. APA 7th Compliance: Citations formatted correctly?
4. Clarity: Is language precise and academic?
5. Coherence: Do paragraphs connect logically?

Provide specific feedback with line references.

"Create a search log in @01_search_and_screening/search_log.md
documenting:
- Databases used: Web of Science, Scopus, etc.
- Search strings and boolean operators
- Date ranges
- Total hits per database"

"Apply /screen to each abstract in
@01_search_and_screening/abstracts.txt
Output to 01_search_and_screening/screening_results.csv with columns:
ID, Title, Authors, Year, Decision, Reason, Confidence"

Using the search log, record ledger, screening results, retrieval ledger
and reviewed report-to-study mapping, reconcile counts at each PRISMA stage.
Keep records, reports and studies separate. Show duplicate removals,
reports not retrieved, full-text exclusions with reasons, included reports
and included studies. Leave UNCERTAIN cases unresolved for human review.

Stop and list missing evidence or inconsistent totals. Do not infer
unobserved stages from the screening CSV or fill a diagram with guessed counts.

For each report with human-confirmed full-text inclusion in the retrieval ledger:
/extract @00_literature_files/[filename]

"Combine all extraction outputs into a single master table
at @02_data_extraction/master_extraction_table.md"

"Review @02_data_extraction/master_extraction_table.md
for missing data, inconsistencies, or errors. Flag any papers
needing manual review."

"Analyze @02_data_extraction/master_extraction_table.md
and identify 5-7 major themes. For each theme, list:
- Papers that discuss it
- Key arguments/findings
- Points of agreement/disagreement"

"Create a methods summary table showing distribution of:
- Qualitative vs quantitative approaches
- Sample sizes and contexts
- Data collection methods
- Analysis techniques"

"Based on all papers, identify:
- Methodological gaps
- Theoretical gaps
- Empirical/contextual gaps
- Temporal gaps

For each gap, explain significance and future research implications."

"Using themes from @03_synthesis_and_writing/themes.md,
draft a literature review section. Structure:
- Introduction to theme
- Discussion of each paper's contribution
- Synthesis of consensus and contradictions
- Implications for research"

/critique @04_manuscript_drafts/literature_section_v1.md

"Extract all citations from @04_manuscript_drafts/full_draft.md
and resolve them against the reviewed Zotero/Better BibTeX bibliography.
Report missing keys; format only verified entries in APA 7th style."

/screen ID: S002; Title: [title]; Authors: [authors]; Year: [year]; Abstract: [abstract]
Decision: UNCERTAIN
→ Add to manual review queue

"For each paper in @00_literature_files/, resolve its DOI to an
OpenAlex ID. Record unresolved DOIs rather than inventing matches.
Fetch referenced_works and citing works (filter=cites:) through cursor
exhaustion. Label edge direction and seed ID; deduplicate candidate IDs
and repeated edges. List candidates connected to at least three distinct
seeds, not three responses or pages, and screen them under our protocol."

"Compare our paper list in @00_literature_files/ with the paper records
exported from ResearchRabbit. Match by DOI or title and list records
not yet in our library. Do not infer citation edges from a paper list;
resolve those through OpenAlex if we need citation-path analysis."

"From all papers in @00_literature_files/ with quantitative results,
extract effect sizes, sample sizes, and statistical significance.
Format as CSV for meta-analysis."

"Ask 3 agents to independently evaluate @00_literature_files/smith2024a.md:
- Agent 1: Assess methodology rigor
- Agent 2: Evaluate theoretical contribution
- Agent 3: Identify practical implications

Keep the stable record/report ID and each assessment’s evidence locators.
Synthesize the assessments, flag disagreements and leave final judgments to me."

"Split the abstracts in @01_search_and_screening/batch1.txt
into 3 groups. Ask an agent to screen each group in parallel
with /screen command. Preserve IDs and source references. Compile results
without overwriting another agent’s file. Leave UNCERTAIN cases for human review;
do not pass them into extraction as final inclusions."

"Reconcile the PRISMA counts from:
- @01_search_and_screening/search_log.md
- @01_search_and_screening/record_ledger.csv
- @01_search_and_screening/screening_results.csv
- @01_search_and_screening/retrieval_ledger.csv
- @01_search_and_screening/study_mapping.csv

Stop on missing evidence, unresolved cases or inconsistent totals.
Do not equate included reports with included studies. When reconciled,
output the stage counts as a Markdown table and a Mermaid diagram;
keep the reviewer-approved ledger revisions with the diagram."

"Convert @02_data_extraction/master_extraction_table.md
to CSV format for:
1. Import into Excel/R for analysis
2. Appendix for manuscript submission
3. Supplementary materials

Preserve one row per report with stable ID, citation key, source filename
and per-field evidence locators. Quote CSV fields and double embedded quotes.
Do not merge different reports solely because their author-year labels match."

"Compile these sections into a complete draft:
- @04_manuscript_drafts/01_introduction.md
- @04_manuscript_drafts/02_methods.md
- @04_manuscript_drafts/03_results.md
- @04_manuscript_drafts/04_discussion.md

Add section numbers and format headers. Resolve citations against the
reviewed bibliography file; report missing keys instead of generating sources."

### AI-Assisted Analysis

[ACTUAL TASKS] were assisted by [CLIENT, PROVIDER, MODEL/VERSION] on
[DATES], using [PROMPT/PROTOCOL REVISION]. [REVIEWER] checked [ACTUAL
COVERAGE AND PROCEDURE]; [UNCHECKED OR UNRESOLVED SCOPE] remains.
The retained prompts, inputs, decisions and verification records are
[ACTUAL LOCATION/AVAILABILITY]. Describe only checks that were performed.

# Export from database to text file
# Format: One record per line, with stable ID, title, authors, year, and abstract

"Process @01_search_and_screening/web_of_science_export.txt:

For each abstract:
1. Apply /screen command
2. Log decision with ID
3. Track exclusion reasons
4. Flag uncertain cases for manual review

Output to 01_search_and_screening/screening_results.csv with columns:
ID, Title, Authors, Year, Decision, Reason, Confidence"

"Show me all abstracts marked 'UNCERTAIN' in 01_search_and_screening/screening_results.csv
for manual review"

"Analyze 01_search_and_screening/screening_results.csv:
- Total abstracts screened
- Inclusion rate (%)
- Top 3 exclusion reasons
- Uncertain cases requiring manual review

Format as summary table."

"This SLR examines organizational scaling and scalability.
Inclusion: Peer-reviewed, 2020-2025, empirical studies on
business scaling. Exclusion: Non-English, opinion pieces,
case studies <3 companies."

/screen ID: S001; Title: Fictional scaling study; Authors: Example only;
Year: 2024; Abstract: A multi-method study describes scaling across
50 firms. The supplied abstract does not establish peer-review status.

→ UNCERTAIN: ask the reviewer to verify the missing eligibility evidence.

/extract @00_literature_files/example_s001.md

→ Generated structured table with RQ, methods, findings, contribution

"Analyze all papers with 'scalability' construct.
Compare definitions across studies. Identify conceptual confusion
vs. conceptual clarity. Draft 1-page synthesis."