philosophy-research

Philosophy Research Skill

Search and retrieve academic philosophy literature using structured APIs instead of WebSearch.

Setup check: If the current directory has no .phillit/ marker, PhilLit has not been set up here — offer to run /phillit:setup (a quick one-time step) before continuing.

When to Use This Skill

Use this skill when:

• Searching for philosophy papers, articles, or concepts
• Accessing SEP (Stanford Encyclopedia of Philosophy) content
• Finding citations and references for papers
• Verifying that a paper exists with correct metadata
• Building bibliographies for literature reviews

Search Workflow

Phase 1: SEP & IEP (Most Authoritative)

# Discover relevant SEP articles
bash "$PHILLIT_ROOT/bin/phillit-run" skills/philosophy-research/scripts/search_sep.py "free will"

# Extract structured content from an SEP entry
bash "$PHILLIT_ROOT/bin/phillit-run" skills/philosophy-research/scripts/fetch_sep.py freewill --sections "preamble,1,2,bibliography"
bash "$PHILLIT_ROOT/bin/phillit-run" skills/philosophy-research/scripts/fetch_sep.py freewill --bibliography-only

# Discover relevant IEP articles (different coverage from SEP)
bash "$PHILLIT_ROOT/bin/phillit-run" skills/philosophy-research/scripts/search_iep.py "free will"

# Extract structured content from an IEP entry
bash "$PHILLIT_ROOT/bin/phillit-run" skills/philosophy-research/scripts/fetch_iep.py freewill --sections "1,2,3,bibliography"
bash "$PHILLIT_ROOT/bin/phillit-run" skills/philosophy-research/scripts/fetch_iep.py freewill --bibliography-only

Parse the bibliographies for foundational works, then search for those papers.

Phase 2: PhilPapers

# Philosophy-specific paper search
bash "$PHILLIT_ROOT/bin/phillit-run" skills/philosophy-research/scripts/search_philpapers.py "epistemic injustice"
bash "$PHILLIT_ROOT/bin/phillit-run" skills/philosophy-research/scripts/search_philpapers.py "virtue epistemology" --recent

Phase 3: Extended Academic Search

# Semantic Scholar - broad academic search
bash "$PHILLIT_ROOT/bin/phillit-run" skills/philosophy-research/scripts/s2_search.py "moral responsibility" --field Philosophy --year 2015-2025

# OpenAlex - 250M+ works, cross-disciplinary
bash "$PHILLIT_ROOT/bin/phillit-run" skills/philosophy-research/scripts/search_openalex.py "consciousness" --year 2020-2024 --min-citations 10

# CORE - 431M+ research outputs, 46M full texts, excellent for abstracts
bash "$PHILLIT_ROOT/bin/phillit-run" skills/philosophy-research/scripts/search_core.py "epistemic injustice" --year 2020-2024
bash "$PHILLIT_ROOT/bin/phillit-run" skills/philosophy-research/scripts/search_core.py --doi "10.1111/nous.12191"
bash "$PHILLIT_ROOT/bin/phillit-run" skills/philosophy-research/scripts/search_core.py --title "Freedom of the Will" --author "Frankfurt"

# arXiv - preprints, AI ethics, recent work
bash "$PHILLIT_ROOT/bin/phillit-run" skills/philosophy-research/scripts/search_arxiv.py "AI alignment ethics" --category cs.AI --recent

Note: CORE API is rate-limited to 5 requests per 10 seconds (free tier, no API key required).

Phase 4: Citation Traversal

# Get references and citations for a paper
bash "$PHILLIT_ROOT/bin/phillit-run" skills/philosophy-research/scripts/s2_citations.py PAPER_ID --both --influential-only

# Find recommendations based on seed papers
bash "$PHILLIT_ROOT/bin/phillit-run" skills/philosophy-research/scripts/s2_recommend.py --positive "PAPER_ID1,PAPER_ID2"

Phase 5: Batch Details

# Efficiently fetch metadata for multiple papers
bash "$PHILLIT_ROOT/bin/phillit-run" skills/philosophy-research/scripts/s2_batch.py --ids "PAPER_ID1,PAPER_ID2,DOI:10.xxx/yyy"

Phase 6: Content Enrichment

Enrich entries that lack abstracts. Two sub-steps:

6a. Abstract Resolution — Resolve abstracts from academic APIs (S2 → OpenAlex → CORE fallback):

# Single paper
bash "$PHILLIT_ROOT/bin/phillit-run" skills/philosophy-research/scripts/get_abstract.py --doi "10.1111/nous.12191"
bash "$PHILLIT_ROOT/bin/phillit-run" skills/philosophy-research/scripts/get_abstract.py --title "Freedom of the Will" --author "Frankfurt" --year 1971
bash "$PHILLIT_ROOT/bin/phillit-run" skills/philosophy-research/scripts/get_abstract.py --s2-id "abc123def"

For batch processing, use enrich_bibliography.py (in literature-review scripts):

bash "$PHILLIT_ROOT/bin/phillit-run" skills/literature-review/scripts/enrich_bibliography.py input.bib

6b. NDPR Enrichment — For @book entries that still lack an abstract after step 6a and have High or Medium importance (as noted in the keywords field), use NDPR (Notre Dame Philosophical Reviews) to extract opening summary paragraphs from book reviews:

# Discover NDPR reviews for a book
bash "$PHILLIT_ROOT/bin/phillit-run" skills/philosophy-research/scripts/search_ndpr.py "book title or author"

# Extract content from an NDPR review
bash "$PHILLIT_ROOT/bin/phillit-run" skills/philosophy-research/scripts/fetch_ndpr.py REVIEW_URL

NDPR extracts are tagged with abstract_source = {ndpr} in BibTeX entries. Note: NDPR content is primarily descriptive of the book's arguments but may include reviewer evaluation.

The batch enrich_bibliography.py script handles both 6a and 6b automatically.

Phase 7: Encyclopedia Context Extraction (Optional)

For important papers, extract how they're discussed in authoritative philosophy encyclopedias:

# Extract SEP citation context for a specific paper
bash "$PHILLIT_ROOT/bin/phillit-run" skills/philosophy-research/scripts/get_sep_context.py freewill --author "Frankfurt" --year 1971
bash "$PHILLIT_ROOT/bin/phillit-run" skills/philosophy-research/scripts/get_sep_context.py freewill --author "Fischer" --year 1998 --coauthor "Ravizza"

# Extract IEP citation context
bash "$PHILLIT_ROOT/bin/phillit-run" skills/philosophy-research/scripts/get_iep_context.py freewill --author "Frankfurt" --year 1971

Phase 8: Verification

# Verify a paper exists via CrossRef
bash "$PHILLIT_ROOT/bin/phillit-run" skills/philosophy-research/scripts/verify_paper.py --title "The Title" --author "Smith" --year 2020
bash "$PHILLIT_ROOT/bin/phillit-run" skills/philosophy-research/scripts/verify_paper.py --doi "10.1093/mind/fzv123"

SEP Content Access

Use fetch_sep.py instead of WebFetch for SEP articles.

fetch_sep.py provides structured extraction:

• Preamble (abstract/introduction)
• Individual sections by number
• Bibliography with parsed author/year/title
• Author and publication dates

# Get specific sections
bash "$PHILLIT_ROOT/bin/phillit-run" skills/philosophy-research/scripts/fetch_sep.py compatibilism --sections "preamble,1,2"

# Get bibliography only
bash "$PHILLIT_ROOT/bin/phillit-run" skills/philosophy-research/scripts/fetch_sep.py freewill --bibliography-only

IEP Content Access

Use fetch_iep.py for Internet Encyclopedia of Philosophy articles.

fetch_iep.py provides structured extraction similar to SEP:

• Preamble, individual sections, bibliography, author information

bash "$PHILLIT_ROOT/bin/phillit-run" skills/philosophy-research/scripts/fetch_iep.py compatibilism --sections "1,2,3"
bash "$PHILLIT_ROOT/bin/phillit-run" skills/philosophy-research/scripts/fetch_iep.py freewill --bibliography-only

Note: IEP has different coverage than SEP. Use both for comprehensive encyclopedia coverage.

Available Scripts

Script	Purpose	Key Options
s2_search.py	Paper discovery	--bulk, --year, --field, --min-citations
s2_citations.py	Citation traversal	--references, --citations, --influential-only
s2_batch.py	Batch paper details	--ids, --file
s2_recommend.py	Find similar papers	--positive, --negative, --for-paper
search_openalex.py	Broad academic search	--year, --doi, --id, --cites, --oa-only
search_arxiv.py	arXiv preprints	--category, --author, --recent, --id
search_sep.py	SEP discovery	--limit, --all-pages
fetch_sep.py	SEP content extraction	--sections, --bibliography-only, --related-only
search_iep.py	IEP discovery	--limit, --all-pages
fetch_iep.py	IEP content extraction	--sections, --bibliography-only
search_philpapers.py	PhilPapers search	--limit, --recent
verify_paper.py	DOI verification	--title, --author, --year, --doi
search_core.py	CORE API (431M papers)	--doi, --title, --author, --year
get_abstract.py	Multi-source abstract resolution	--doi, --s2-id, --title, --author
get_sep_context.py	SEP citation context extraction	--author, --year, --coauthor
get_iep_context.py	IEP citation context extraction	--author, --year, --coauthor
search_ndpr.py	NDPR book review discovery	--limit
fetch_ndpr.py	NDPR content extraction	--sections
check_setup.py	Environment check	(no options)

Output Format

All scripts output JSON with consistent structure:

{
  "status": "success|partial|error",
  "source": "script_name",
  "query": "search query",
  "results": [...],
  "count": 5,
  "errors": []
}

Exit codes: 0=success, 1=not found, 2=config error, 3=API error

WebFetch Usage

Use WebFetch only when scripts don't provide the needed content:

• PhilPapers entry pages for additional metadata
• Publisher pages for paper details not in S2
• DOI resolution verification (https://doi.org/{doi})

Do NOT use WebFetch for:

• SEP articles: use fetch_sep.py (structured output)
• Paper abstracts: use s2_search.py or s2_batch.py

Verification Requirements

• Never fabricate citations - only include papers found via scripts
• Verify DOIs via verify_paper.py when S2 lacks DOI
• Omit DOI field if verification fails (never invent DOIs)
• Report gaps if expected papers are not found

Environment

Required environment variables (set in .env at the project root — takes priority over shell environment):

BRAVE_API_KEY        # Required for SEP/PhilPapers discovery
CROSSREF_MAILTO      # Required for CrossRef polite pool
S2_API_KEY           # Recommended for Semantic Scholar
OPENALEX_EMAIL       # Recommended for OpenAlex polite pool

Check setup with:

bash "$PHILLIT_ROOT/bin/phillit-run" skills/philosophy-research/scripts/check_setup.py