Pharma Skill System
PassAudited by ClawScan on May 1, 2026.
Overview
This is a coherent instruction-only pharma workflow skill, but it may handle sensitive medical, customer, compliance, and business data, so users should control what information they provide and where outputs are stored.
This skill appears safe to install as an instruction-only workflow library, but use it carefully with pharma data. Do not paste patient-identifiable information, confidential business plans, CRM exports, budgets, contracts, or PV records unless your organization allows that use. Keep human review for compliance, medical, regulatory, publication, and financial outputs.
Findings (4)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Confidential medical, commercial, training, or compliance information could be reused in later answers if placed into a shared knowledge base without controls.
The skill explicitly includes building and maintaining a company knowledge base and Q&A interface, which can persist and reuse internal information across tasks.
构建和维护公司动态知识库，提供自然语言问答接口，自动生成培训材料，规划个性化学习路径，评估培训效果。
Use approved storage locations, restrict access to the knowledge base, review source quality, and avoid adding confidential or personal data unless the retention and sharing rules are clear.
Patient-safety or adverse-event information may contain personal or regulated health data and should not be casually copied into an AI workflow.
The compliance/PV workflow can process adverse-event reports and ICSR records, which are sensitive regulated medical-safety data.
C6-15 ICSR处理全流程SOP - 输入: (必选)不良事件报告 ... 输出: ICSR处理记录（接收→评估→随访→报告→归档全流程，时效追踪）
Provide only the minimum necessary data, de-identify patient information where possible, and have PV/compliance personnel verify any generated records or reports before use.
If connected to browsing, scraping, or enterprise data tools, this workflow could collect more information than intended unless sources and limits are specified.
The intelligence cluster describes continuous monitoring and collection from public and internal information sources, though no code, endpoints, or credentials are provided.
7x24小时监控内外部信息源（市场、竞品、政策、医学文献、社媒），实时抓取、结构化、分类、初步分析
Before using monitoring workflows, define allowed sources, time windows, collection limits, and review steps, especially for internal or social-media data.
Incorrect data or AI-generated assumptions could spread from one workflow into downstream business, regulatory, or compliance materials.
The skill intentionally chains outputs across many business clusters, so a mistaken or unreviewed output can influence later reports, plans, or compliance drafts.
一个单元的输出可作为另一集群单元的输入，按组合接口串联
Treat generated outputs as drafts, validate key facts before using them as inputs to later pipeline steps, and require human review for regulated or high-impact decisions.