Security audit

保险理赔维权辅助分析

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent insurance claim appeal drafting skill, but users should handle the requested identity, policy, and medical information carefully.

Install only if you want Chinese insurance denied-claim appeal assistance. Before using it, redact unnecessary ID digits, contact details, unrelated medical history, and policy identifiers where possible; share only the documents needed for the specific dispute and use official secure insurer or regulator channels for submission.

A.I.G

Vulnerability Patterns

Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (13)

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 90% confidence
Finding: The trigger list includes several broad, everyday phrases such as “拒赔”, “理赔纠纷”, and “怎么应对拒赔”, which can activate the skill in conversations that are only loosely related to formal insurance-claim appeal workflows. Because this skill solicits sensitive financial, insurance, and medical details, accidental invocation increases the risk of unnecessary collection or exposure of highly sensitive personal information.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The skill explicitly requests policy numbers, claim amounts, medical diagnoses, treatment history, and pre-underwriting health disclosures, but does not present an upfront privacy warning, data-minimization guidance, or redaction instructions. In this legal-finance and health context, collecting such data without clear sensitivity notices materially raises the risk of oversharing regulated or highly sensitive personal information.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The template explicitly solicits highly sensitive personal data, including identity number and contact details, in a document used for insurance disputes, but provides no warning about minimization, redaction, secure transmission, or retention. In this context, users may copy, store, or share the completed letter insecurely, increasing the risk of privacy leakage, identity abuse, and unnecessary exposure of regulated personal and health-related information.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 96% confidence
Finding: The evidence checklist requests medical records, diagnostic proofs, and test reports, which are highly sensitive health documents, yet the template gives no warning about privacy, over-disclosure, or the risks of submitting more medical information than necessary. Because this skill is specifically for insurance claim disputes, users are likely to provide exactly these materials, making inadvertent exposure of medical data a realistic and foreseeable harm.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 92% confidence
Finding: The template explicitly solicits highly sensitive personal data, including full name, national ID number, phone number, address, and insurer details, but provides no minimization, redaction, storage, or transmission guidance. In a claims-dispute workflow, users may copy, share, or submit this document through insecure channels, increasing the risk of identity theft, privacy violations, or unnecessary exposure of regulated personal information.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 92% confidence
Finding: This section instructs users to collect highly sensitive medical records, test results, diagnoses, and insurance-related documents, but provides no privacy, minimization, retention, or secure-handling guidance. In a dispute-support workflow, this increases the risk of over-collection, improper sharing, and exposure of health data and other personal information beyond what is necessary for the appeal.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: This section asks for emails, detailed phone communication records, and call recordings, but does not warn about privacy, consent, or jurisdiction-specific recording rules. That can lead users to gather or create recordings unlawfully, retain excessive personal data, or share communications in ways that create legal and privacy exposure for both the customer and third parties.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 91% confidence
Finding: The trigger phrase at line 9 is short and generic within the insurance domain, making accidental invocation more likely during ordinary discussion of claim disputes. In a legal-finance support skill, unintended activation can route users into specialized advice flows without clear intent, creating reliability and compliance risks.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 87% confidence
Finding: The trigger phrase at line 10 is broad enough to match common conversations about insurance claims rather than a deliberate request to use this skill. Because the skill operates in a sensitive legal/financial context, unintended invocation can misclassify user needs and surface quasi-legal guidance when the user may only be asking a general question.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 84% confidence
Finding: The trigger phrase at line 17 is ambiguous and likely to overlap with routine insurance discussions, which increases the chance of accidental skill activation. Given this skill produces complaint letters and rights-protection workflows, accidental entry into the workflow could lead to inappropriate or overconfident assistance in a regulated domain.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 89% confidence
Finding: The file title and all instructional content are written exclusively in Chinese, with no indication that users may choose another language or that the skill is limited to a Chinese-speaking or China-specific deployment context. Under the stated policy, forcing a specific language without opt-in is a natural-language policy violation unless the locale restriction is explicit and justified.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 88% confidence
Finding: The entire compliance document is written as mandatory output guidance in Chinese, including required disclaimer text to append on every response. There is no indication that users may choose another language or that the skill is restricted to a Chinese-speaking or region-specific context, which creates a natural-language locale policy concern.
Content: No source excerpt is available for this finding.

Overly Broad Trigger

Low

Category: Trigger Abuse
Confidence: 84% confidence
Finding: Using the single short trigger “拒赔” is overly generic and may match many unrelated discussions about denied claims, reimbursements, or general complaints. In this skill, unintended activation is more concerning because the workflow immediately moves toward gathering sensitive legal, insurance, and medical case information.
Content: No source excerpt is available for this finding.