tooluniverse-pharmacovigilance
Analyze drug safety signals from FDA adverse event reports, label warnings, and pharmacogenomic data. Calculates disproportionality measures (PRR, ROR), identifies serious adverse events, assesses pharmacogenomic risk variants. Use when asked about drug safety, adverse events, post-market surveillance, or risk-benefit assessment.
npx skills add BioTender-max/awesome-bio-agent-skills --skill tooluniverse-pharmacovigilance --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
What it does
Analyzes drug safety signals using FAERS adverse event data, FDA labeling information, PharmGKB pharmacogenomics annotations, and clinical trial safety signals. Produces disproportionality measures (PRR, ROR), identifies serious adverse events, and evaluates pharmacogenomic risk variants to inform risk-benefit assessments and recommendations.
How it works
Follows a multi-phase workflow:
- Drug identification: resolves drug identity to generic name, retrieves identifiers, and class/mechanism.
- Adverse Event Profiling (FAERS): queries FAERS for drug-event pairs, collects counts, and computes disproportionality measures (PRR, ROR); stratifies by seriousness; outputs a ranked AE table.
- Label Warning Extraction: fetches DailyMed label sections (boxed warnings, contraindications, warnings/precautions, adverse reactions, interactions) and categorizes warning severities.
- Pharmacogenomic Risk: queries PharmGKB for PGx annotations, compiles clinically actionable variants with recommended actions and evidence level.
- Clinical Trial Safety: gathers Phase 3 safety data from ClinicalTrials.gov and summarizes serious AEs and deaths from trials.
- Pathway & Mechanism Context (NEW): gathers KEGG metabolism pathways and target pathways to contextualize safety mechanisms.
- Literature Intelligence (ENHANCED): pulls PubMed, BioRxiv, MedRxiv, and OpenAlex sources for safety evidence.
- Signal Prioritization: ranks signals by PRR, severity, and frequency, and identifies actionable signals with risk-benefit considerations.
- Report Synthesis: compiles outputs into reports and separate data files for adverse events and pharmacogenomics.
When to use it
Use when asked about drug safety, adverse events, post-market surveillance, or risk-benefit assessment for a specific drug (e.g., safety concerns, associated adverse events, comparisons of safety profiles, or pharmacovigilance analyses).
What it can touch
- Tools referenced include: DailyMed_search_spls, DailyMed_get_spl_by_set_id, FAERS_count_reactions_by_drug_event, FAERS_get_event_details, PharmGKB_search_drug, ChEMBL_search_drugs, DailyMed_get_spl_by_set_id, DailyMed_search_spls, OpenFDA_get_drug_events, DailyMed_get_spl_by_set_id, ClinicalTrials.gov related tools, PubMed_search_articles, BioRxiv_search_preprints, MedRxiv_search_preprints, kegg_search_pathway, kegg_get_gene_info, DailyMed_get_spl_by_set_id.
- Outputs include:
[DRUG]_safety_report.md,[DRUG]_adverse_events.csv,[DRUG]_pharmacogenomics.csv.
Caveats
- Requires explicit source citations for every safety signal following the mandated formatting.
- Uses a report-first approach: create the report file first with placeholders, then progressively update as data is gathered.
- English-first queries: drug names and search terms should be in English for tool calls; original-language terms may be used as fallback.
# Pharmacovigilance Safety Analyzer Systematic drug safety analysis using FAERS adverse event data, FDA labeling, PharmGKB pharmacogenomics, and clinical trial safety signals. **KEY PRINCIPLES**: 1. **Report-first approach** - Create report file FIRST, update progressively 2. **Signal quantification** - Use disproportionality measures (PRR, ROR) 3. **Severity stratification** - Prioritize serious/fatal events 4. **Multi-source triangulation** - FAERS, labels, trials, literature 5. **Pharmacogenomic context** - Include genetic risk factors 6. **Actionable output** - Risk-benefit summary with recommendations 7. **English-first queries** - Always use English drug names and search terms in tool calls, even if the user writes in another language. Only try original-language terms as a fallback. Respond in the user's language --- ## When to Use Apply when user asks: - "What are the safety concerns for [drug]?" - "What adverse events are associated with [drug]?" - "Is [drug] safe? What are the risks?" - "Should I be concerned about [specific adverse event] with [drug]?" - "Compare safety profiles of [drug A] vs [drug B]" - "Pharmacovigilance analysis for [drug]" --- ## Critical Workflow Re
- When to Use
- Critical Workflow Requirements
- 1. Report-First Approach (MANDATORY)
- 2. Citation Requirements (MANDATORY)
- Phase 0: Tool Verification
- Known Parameter Corrections
- Workflow Overview
- Phase 1: Drug Disambiguation
- 1.1 Resolve Drug Identity
- 1.2 Output for Report
- Phase 2: Adverse Event Profiling
- 2.1 FAERS Query Strategy
- 2.2 Disproportionality Analysis
- 2.3 Severity Classification
What does the tooluniverse-pharmacovigilance skill do?
Analyze drug safety signals from FDA adverse event reports, label warnings, and pharmacogenomic data. Calculates disproportionality measures (PRR, ROR), identifies serious adverse events, assesses pharmacogenomic risk variants. Use when asked about drug safety, adverse events, post-market surveillance, or risk-benefit assessment.
How do I install it?
Run `npx skills add BioTender-max/awesome-bio-agent-skills --skill tooluniverse-pharmacovigilance --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From BioTender-max/awesome-bio-agent-skills, a repository with 135 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.