菜单
Conduct PRISMA 2020-compliant systematic reviews using Azure AI services. Covers protocol development, search strategy, screening, and reporting.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Microsoft Identity and Access Administrator (SC-300) grounded skill. Covers Entra ID tenant management, user/group/device identities, hybrid identity, authentication methods (MFA, passkeys, FIDO2, certificate-based), Conditional Access, ID Protection, Global Secure Access, workload identities (managed identities, service principals, app registrations), enterprise app integration, identity governance (PIM, access reviews, entitlement management), and monitoring (KQL, workbooks, Identity Secure Score). Use when configuring Entra ID, managing authentication, setting up Conditional Access, implementing managed identities, reviewing access, or troubleshooting identity issues.
Master skill bundle for InsightPulseAI. Covers Azure resource constants (ipai-resource-map), Odoo 18 CE development conventions (ipai-odoo-platform), Pulser agent platform patterns (ipai-agent-platform), and Odoo docs enhancement. Load for any IPAI Azure, Odoo, Foundry, or agent work. Triggers on: IPAI, InsightPulseAI, Pulser, Odoo 18 CE, ipai_*, BIR compliance, ACA deployment, Foundry, rg-ipai, pg-ipai-odoo, ipai-copilot-resource.
Deploy and configure Microsoft's SaaS Accelerator for marketplace offers. Pre-built landing page, fulfillment webhook, subscription management — eliminates custom development for marketplace technical integration.
End-to-end guide for publishing a SaaS offer on Microsoft AppSource/Azure Marketplace. Covers Partner Center setup, offer creation, technical integration, listing content, and certification. Based on Microsoft's Mastering the Marketplace curriculum.
Extract structured data from research papers using Azure Document Intelligence and GPT-4.1 vision. Automates PICO extraction, risk of bias assessment, and data table population.
Perform meta-analyses using Azure Databricks for computation and Power BI for visualization. Supports fixed/random effects, subgroup analysis, and publication bias assessment.
| name | prismalab-systematic-review |
| description | Conduct PRISMA 2020-compliant systematic reviews using Azure AI services. Covers protocol development, search strategy, screening, and reporting. |
ipai-copilot-resource): GPT-4.1 for protocol drafting, search strategy generationsrch-ipai-dev-sea, index: prismalab-rag-v1): RAG over methodology corpusGenerate a PROSPERO-ready protocol from a research question:
Input: "What is the effectiveness of AI-assisted diagnosis in diabetic retinopathy?"
Output: Structured protocol with PICO, objectives, search strategy, eligibility criteria
Generate database-specific search strings:
# PubMed search string generation
search = {
"population": "diabetic retinopathy patients",
"intervention": "artificial intelligence OR deep learning OR machine learning",
"comparator": "ophthalmologist OR human expert",
"outcome": "diagnostic accuracy OR sensitivity OR specificity",
"filters": ["humans", "english", "2019-2026"]
}
Use Boolean operators adapted per database:
Two-phase screening approach:
Generate the PRISMA 2020 flow diagram data:
Records identified (n=X)
→ Duplicates removed (n=X)
→ Screened (n=X)
→ Excluded (n=X)
→ Full-text assessed (n=X)
→ Excluded with reasons (n=X)
→ Included in synthesis (n=X)
→ Quantitative synthesis (n=X)