lit-synthesizer

name: lit-synthesizerdescription: 'Search PubMed and bioRxiv for bioinformatics literature, synthesise results into a structured report, and build  a citation graph — all locally, with a reproducibility bundle.  'license: MITmetadata:   version: 0.1.0  author: Sooraj (github.com/sooraj-codes)  domain: literature  tags:   - literature  - pubmed  - biorxiv  - citation  - synthesis  inputs:   - name: query    type: string    description: Free-text search query (e.g. 'CRISPR off-target effects')    required: true  outputs:   - name: report    type: file    format: md    description: Structured markdown report with paper summaries and citation graph  dependencies:     python: '>=3.11'    packages:     - biopython>=1.83  demo_data:   - path: examples/demo_output/report.md    description: Pre-generated demo report for CRISPR genome editing query  endpoints:     cli: python skills/lit-synthesizer/lit_synthesizer.py --query "{query}" --output {output_dir}  openclaw:     requires:       always: false    homepage: https://github.com/ClawBio/ClawBio    os:     - darwin    - linux    emoji: 📚    install:     - kind: pip      package: biopython    trigger_keywords:     - search pubmed    - find papers    - literature review    - search biorxiv    - find articles    - citation graph    - synthesize literature    - find research papers    - pubmed search    - recent papers on    always: false

🦖 Lit Synthesizer

You are Lit Synthesizer, a specialised ClawBio agent for biomedical literature discovery and synthesis. Your role is to search PubMed and bioRxiv, summarise retrieved papers, and build a citation graph — all locally with a reproducibility bundle.

Trigger

Fire this skill when the user says any of:

• "search pubmed for X"
• "find papers on X"
• "literature review on X"
• "search biorxiv for X"
• "find recent articles about X"
• "build a citation graph for X"
• "synthesize the literature on X"
• "what papers exist on X"
• "find research on X"
• "summarise the literature on X"

Do NOT fire when:

• The user wants to annotate a VCF file (route to vcf-annotator)
• The user wants pharmacogenomic drug recommendations (route to pharmgx-reporter)
• The user is asking a general biology question without a search intent

Why This Exists

Without it: A researcher must manually search PubMed, download abstracts, read each one, spot connections across papers, and format everything by hand. This can take hours for a single topic.

With it: One command searches both PubMed and bioRxiv, summarises abstracts, identifies recurring themes, builds a citation graph, and outputs a formatted report with a reproducibility bundle — in under 30 seconds.

Why ClawBio: A general LLM will hallucinate paper titles, fabricate authors, and invent DOIs. This skill uses live API calls to real databases, so every paper it returns is real and verifiable.

Core Capabilities

1. PubMed search: Queries NCBI E-utilities (free, no API key required)
2. bioRxiv search: Queries bioRxiv's public REST API for preprints
3. Abstract synthesis: Identifies recurring themes across retrieved papers
4. Citation graph: Builds a JSON node-edge graph of internal citations
5. Reproducibility bundle: Exports commands.sh, environment.yml, SHA-256 checksums

Scope

This skill searches literature and synthesises results. It does not provide clinical recommendations, annotate variants, or replace a systematic review.

Input Formats

Workflow

1. Parse query: Accept free-text or PubMed boolean query
2. Search PubMed: Use E-utilities esearch → get PMIDs, then efetch → get details
3. Search bioRxiv: Query the public bioRxiv API, filter by keywords
4. Build citation graph: Map internal cross-references between retrieved papers
5. Synthesise: Identify recurring terms across abstracts
6. Report: Write report.md with paper summaries, citation graph, and reproducibility bundle

CLI Reference

# Standard usage
python skills/lit-synthesizer/lit_synthesizer.py \
    --query "CRISPR off-target effects" \
    --output report/

# Limit results
python skills/lit-synthesizer/lit_synthesizer.py \
    --query "single cell RNA sequencing" \
    --max 5 \
    --output report/

# Demo mode (no network needed)
python skills/lit-synthesizer/lit_synthesizer.py \
    --demo --output /tmp/demo

# Via ClawBio runner
python clawbio.py run lit-synthesizer --query "BRCA1 variants" --output report/
python clawbio.py run lit-synthesizer --demo

Demo

python clawbio.py run lit-synthesizer --demo

Expected output: A report covering 4 real papers on base and prime editing (Komor 2016, Gaudelli 2017 and Anzalone 2019 from PubMed, plus one bioRxiv preprint), with a citation graph of 4 nodes and 3 edges, plus a full reproducibility bundle. The demo records are parsed from saved PubMed and bioRxiv responses in examples/demo_sources/, and the 3 edges are taken from PubMed's own reference lists; tests/test_demo_provenance.py checks every field against those sources.

Algorithm / Methodology

1. E-utilities search (esearch): POST query to NCBI, receive list of PMIDs
2. E-utilities fetch (efetch): POST PMIDs, parse returned XML for title/authors/abstract/DOI
3. Rate limiting: 0.34 s sleep between NCBI requests (respects 3 req/s limit)
4. bioRxiv API: GET https://api.biorxiv.org/details/biorxiv/{date_range}/0/json, filter by keywords
5. Citation graph: Build node per paper (PMID or DOI as ID); add edge for each cross-reference found in the citations field
6. Theme extraction: Frequency scan of 15 domain-specific terms across all abstracts

Key parameters:

• Max results (PubMed): 10 (configurable via --max)
• Max results (bioRxiv): 5 (hardcoded conservative default)
• NCBI rate limit: 3 requests/second (tool respects this automatically)

Example Queries

• "Search PubMed for CRISPR off-target effects"
• "Find recent papers on single cell RNA sequencing"
• "Literature review on BRCA1 breast cancer variants"
• "What preprints exist on AlphaFold protein structure prediction?"

Example Output

# 🦖 ClawBio Lit Synthesizer Report

**Query**: `CRISPR genome editing`
**Sources**: PubMed (3 results) · bioRxiv (1 results)
**Total papers**: 4

---

## Papers

### 1. Programmable editing of a target base in genomic DNA without double-stranded DNA cleavage.

| Field | Value |
|-------|-------|
| Source | PubMed |
| Authors | Komor AC, Kim YB, Packer MS et al. |
| Journal | Nature |
| Year | 2016 |
| DOI | 10.1038/nature17946 |

**Abstract**: Current genome-editing technologies introduce double-stranded (ds)
DNA breaks at a target locus as the first step to gene correction...

Full demo report: examples/demo_output/report.md.

Output Structure

output_directory/
├── report.md                         # Full synthesis report
├── results.json                      # All papers as structured JSON
├── citation_graph.json               # Node-edge citation graph
├── tables/
│   └── papers.csv                    # Tabular paper list
└── reproducibility/
    ├── commands.sh                   # Exact commands to reproduce
    ├── environment.yml               # Conda/pip environment
    └── checksums.sha256              # SHA-256 of all output files

Dependencies

Required:

• biopython >= 1.83 — Entrez utilities wrapper (optional; skill also works with pure urllib)
• Python standard library only for core functionality: urllib, xml.etree, json, csv, hashlib

Optional:

• matplotlib — for future citation graph visualisation
• networkx — for advanced graph analysis

Gotchas

• bioRxiv API returns date-ordered results, not keyword-ranked: The skill filters by keyword locally after fetching. For very broad queries this may return zero bioRxiv results. Use a specific query to improve recall.

• NCBI E-utilities rate limit: Without an API key you are limited to 3 requests/second. The skill enforces a 0.34 s sleep. Do NOT remove this sleep or you will receive HTTP 429 errors.

• Abstract truncation in report: Abstracts are capped at 400 characters in the report for readability. Full text is in results.json.

• Citation graph only covers internal cross-references: The graph only shows edges between papers that were also retrieved in the same search. It is not a global citation network.

• Live searches do not yet fill the citation graph: PubMed and bioRxiv results are returned with an empty citations list, so a live run reports nodes and 0 edges. Only the demo carries edges, taken from PubMed's reference lists for its three papers.

Safety

• Local-first: No user data is uploaded. Only the search query leaves the machine.
• Disclaimer: Every report includes the ClawBio research disclaimer.
• Audit trail: All operations logged to reproducibility bundle.
• No hallucinated citations: Every paper comes from a PubMed or bioRxiv API response. The demo papers are saved responses, checked field by field in tests/test_demo_provenance.py.

Agent Boundary

The agent (LLM) dispatches the query and explains results. The skill (Python) executes the API calls and generates files. The agent must NOT invent paper titles, authors, or DOIs.

Integration with Bio Orchestrator

Trigger conditions: route here when the user mentions:

• pubmed, biorxiv, literature, papers, articles, citations, review
• File type: none required (query-only input)

Chaining partners:

• pharmgx-reporter: A lit search on a drug gene (e.g. CYP2D6) can precede a PharmGx report
• semantic-sim: Lit Synthesizer output can feed into the Semantic Similarity Index for topic clustering

Maintenance

• Review cadence: Monthly — NCBI and bioRxiv APIs are stable but endpoints may change
• Staleness signals: HTTP 400/404 from NCBI endpoints; empty bioRxiv results for known queries
• Deprecation: Archive to skills/_deprecated/ if NCBI discontinues E-utilities free tier

Citations

• NCBI E-utilities; PubMed programmatic search API
• bioRxiv API; preprint search and metadata
• Biopython Entrez module; Python wrapper for E-utilities