Insurance Anti Fraud
PassAudited by VirusTotal on May 11, 2026.
Overview
Type: OpenClaw Skill Name: insurance-anti-fraud Version: 2.0.0 The skill is a specialized tool for insurance anti-fraud analysis, strictly aligned with its stated purpose of assisting claims and risk control teams within the Chinese insurance market. It provides structured domain knowledge, regulatory frameworks (NFRA 2024), and illustrative Python logic for risk scoring and investigation workflows. There is no evidence of malicious intent, data exfiltration, or unauthorized command execution in files like SKILL.md or the reference documents.
Findings (0)
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.
If treated as an automatic decision-maker, the skill could contribute to claim denial, investigation escalation, or reporting decisions without adequate human review.
The skill's workflow includes high-impact claim outcomes and possible criminal referral. The artifacts do not provide tools that execute these actions, but the guidance could influence consequential decisions.
风险分级 ... HIGH_RISK → 深度调查+报案 ... 结论处置 ... 拒赔处理 → 发拒赔通知+说明 ... 报案追究 → 移送公安（涉嫌犯罪）
Use the skill only for triage and analysis support; require qualified human, legal, and compliance approval before denial, blacklisting, reporting, or other adverse action.
Using real claimant data without proper authority or minimization could expose sensitive personal, medical, or financial information and create compliance risk.
The reference material expects access to regulated medical, insurance, and financial records. This is purpose-aligned for insurer anti-fraud work, but it is sensitive and should be limited to authorized contexts.
带病投保 ... 调取体检记录、医保数据 ... 重复索赔 ... 银保信数据查询 ... 高额保额 ... 财务证明核查
Only use data the organization is legally authorized to process, apply least-privilege access, redact unnecessary identifiers, and follow insurance, medical-data, and privacy compliance requirements.
Sensitive case details could be exposed in prompts, transcripts, logs, or any enabled memory features if users paste full claim files unnecessarily.
The documented usage invites users to place claim-case information into the agent context. No persistent memory or storage mechanism is shown, but claim cases may contain sensitive personal data.
/insurance-anti-fraud "Analyze this claim case for fraud risk" /insurance-anti-fraud "Generate fraud risk scoring for these 10 cases"
Provide only the minimum necessary case facts, de-identify data where possible, and ensure any agent memory, logging, or retention settings comply with internal privacy rules.
Outdated, incomplete, or jurisdiction-specific compliance guidance could be relied on too heavily in regulated insurance decisions.
The description uses strong compliance and completeness language. That is not malicious, but users may over-trust legal or regulatory guidance if they do not independently verify it.
AI-powered insurance anti-fraud analysis expert — the definitive skill ... Covers the complete CBIRC Anti-Fraud Framework
Verify cited regulations and workflows against current official sources and internal legal/compliance policies before operational use.