菜单
AI-powered SQL security scanning and auto-fix for Databricks and general SQL. Use when the user asks to scan SQL for injection vulnerabilities, detect PII in databases, validate SQL compliance, analyze database privileges, or auto-fix SQL security issues.
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
Risk-focused architecture analysis that unearths deep insights specific to the codebase. Use when the user asks to analyze architecture, find architectural risks, review system design, verify auth routes, analyze failure modes, generate architecture diagrams, or understand tech stack trade-offs.
Evidence-based compliance auditing for SOC2, HIPAA, PCI-DSS, and ISO 27001 frameworks. Use when the user asks about compliance readiness, audit preparation, control gap analysis, certification requirements, or framework-specific compliance assessments.
Automatic security scan triggered before git commits. Runs the Opsera security scan tool against the entire repo, categorizes findings into new (staged) vs existing, and blocks commits only if staged changes have critical/high issues.
Technical security scanning for vulnerabilities, secrets, SAST, container security, and infrastructure-as-code issues. Use when the user asks for vulnerability scanning, secret detection, code security analysis, container scanning, or IaC security checks.
| name | sql-security |
| description | AI-powered SQL security scanning and auto-fix for Databricks and general SQL. Use when the user asks to scan SQL for injection vulnerabilities, detect PII in databases, validate SQL compliance, analyze database privileges, or auto-fix SQL security issues. |
Perform SQL security analysis using the mcp__opsera__sql-security tool.
sql_file path)table name in catalog.schema.table format)sql_file and optionally compliance_standard)sql_file, sets auto_fix: true)table and/or user)mcp__opsera__sql-security with the collected parameters.mcp__opsera__opsera_report_telemetry with:
toolName: sql-securitystatus: success/partial/failedtarget: the SQL file or table analyzedtargetType: codecategories: type of findings (e.g., "sql_injection,pii,hardcoded_credentials")