Claude-skill-registry alert-system

install
source · Clone the upstream repo
git clone https://github.com/majiayu000/claude-skill-registry
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/majiayu000/claude-skill-registry "$T" && mkdir -p ~/.claude/skills && cp -r "$T/skills/data/alert-system" ~/.claude/skills/majiayu000-claude-skill-registry-alert-system && rm -rf "$T"
manifest: skills/data/alert-system/SKILL.md
safety · automated scan (low risk)
This is a pattern-based risk scan, not a security review. Our crawler flagged:
  • pip install
Always read a skill's source content before installing. Patterns alone don't mean the skill is malicious — but they warrant attention.
source content

Alert System Skill

Automated monitoring and alerting for drug discovery intelligence.

Quick Start

/alerts --create --target "EGFR inhibitors" --events clinical,fda
/alerts --list
/alerts --check my-alert

What's Included

Alert TypeDescriptionSources
Clinical UpdatesTrial status changesClinicalTrials.gov
RegulatoryFDA/EMA decisionsFDA, EMA
PublicationsNew papers on targetPubMed
CompetitorPipeline updatesPress releases
PatentNew patent filingsUSPTO, EPO
ConferenceMeeting abstractsASCO, ESMO

Alert Configuration

name: EGFR Competitive Alerts
description: Monitor EGFR inhibitor landscape

targets:
  - EGFR
  - Exon 19 deletion
  - T790M
  - C797S

competitors:
  - AstraZeneca
  - Johnson & Johnson
  - Roche

events:
  - clinical_trial_updates
  - regulatory_approvals
  - publications
  - patent_filings

filters:
  phase: [2, 3]
  countries: [US, EU, CN, JP]
  minimum_significance: high

notifications:
  email: user@example.com
  frequency: weekly
  format: summary

Output Structure

# Alert Report: EGFR Landscape (Weekly)

## Summary
| Alert Type | New | Total |
|------------|-----|-------|
| Clinical Updates | 3 | 127 |
| Publications | 12 | 456 |
| Regulatory | 1 | 23 |
| Patent Filings | 5 | 89 |

## Clinical Trial Updates

### New Trials
| NCT ID | Sponsor | Phase | Indication | Status |
|--------|---------|-------|------------|--------|
| NCT01234567 | AstraZeneca | 3 | NSCLC | Recruiting |
| NCT01234568 | Johnson & Johnson | 2 | NSCLC | Not recruiting |

### Status Changes
| NCT ID | Previous | Current | Date |
|--------|----------|---------|------|
| NCT07890123 | Recruiting | Active, not recruiting | 2024-12-10 |

## Key Publications

1. **Fourth-generation EGFR inhibitors targeting C797S**
   - Journal: Nature Cancer
   - Date: 2024-12-08
   - Authors: Zhang et al.
   - Impact: Novel scaffold for resistance

2. **Combination therapy: EGFR + MET inhibition**
   - Journal: Lancet Oncology
   - Date: 2024-12-05
   - Authors: Smith et al.
   - Impact: Positive Phase 2 results

## Regulatory Updates

### FDA Actions
| Date | Sponsor | Action | Drug |
|------|---------|--------|------|
| 2024-12-12 | AstraZeneca | Approval | Tagrisso (adjuvant) |

## Patent Filings

### New Patents
| Date | Assignee | Title | Patent No. |
|------|----------|-------|------------|
| 2024-12-10 | Merck | EGFR C797S inhibitors | US2024/XXX |
| 2024-12-08 | Roche | Antibody-drug conjugate | EPXXX |

## Recommendations

**Action Items**:
1. Review new C797S inhibitor paper - assess differentiation
2. Monitor AstraZeneca adjuvant approval impact
3. Investigate Merck's new patent - potential FTO issue

Event Types

EventDescriptionPriority
Clinical trial startsNew trial initiatedMedium
Clinical trial completesPrimary endpoint reachedHigh
Regulatory approvalDrug approvalHigh
Regulatory rejectionCRL, rejectionHigh
Key publicationHigh-impact paperMedium
Competitor dealM&A, licensingHigh
Patent grantedNew patent issuedMedium
Patent expiryPatent expiresLow

Running Scripts

# Create new alert
python scripts/alert_system.py --create my-alert --target "EGFR"

# Check for updates
python scripts/alert_system.py --check my-alert

# List all alerts
python scripts/alert_system.py --list

# Export alert history
python scripts/alert_system.py --export my-alert --format csv

Requirements

pip install requests schedule

# For persistence
pip install sqlalchemy

Reference

Best Practices

  1. Set specific targets: Narrower = fewer false positives
  2. Use appropriate frequency: Daily for high-priority, weekly for others
  3. Review regularly: Alerts need human interpretation
  4. Adjust filters: Refine based on signal-to-noise
  5. Document actions: Track alert-driven decisions

Common Pitfalls

PitfallSolution
Too many alertsNarrow target/event filters
False positivesAdd confidence filters
Alert fatigueAdjust frequency and thresholds
Missing contextInclude summary with links
No action trackingDocument response to alerts

Alert Frequency Guidelines

PriorityCheck FrequencyNotification
CriticalHourlyImmediate
HighDailyDaily digest
MediumWeeklyWeekly summary
LowMonthlyMonthly report