Back to skill

Security audit

Ai Intelligent Fraud Detection

Security checks for vulnerabilities and agentic risk

Overview

The skill matches a fraud-detection purpose, but it asks users to run unverified remote code and gives little control guidance for sensitive financial and identity data.

Review this before installing. Only run it in a restricted virtual environment or container, pin the repository to a reviewed commit, verify dependencies, and avoid providing real identity documents, account records, or transaction histories until the publisher documents data handling and provenance.

Vulnerability Patterns
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • 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
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Findings (1)

T03 · Remote Payload Retrieval and Execution

Error
Location
SKILL.md:39
Finding

Unpinned Remote Repository and Dependency Execution

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The skill description is broad and does not define clear activation triggers, scope limits, or invocation constraints for high-sensitivity functions like transaction monitoring and identity verification. In an agent ecosystem, this can cause the skill to be invoked in unintended contexts, increasing the chance of over-collection, over-processing, or misuse of sensitive financial and identity data.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The markdown omits warnings that the skill handles highly sensitive activities: identity verification and transaction monitoring. Without prominent disclosure and consent language, users may provide personal, financial, or account data without understanding the privacy, compliance, and surveillance implications, which is especially risky in fraud and KYC-style contexts.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
79% confidence
Finding

All user-facing natural-language content in the skill metadata and documentation is presented only in Chinese, with no indication of language choice or opt-in. This may violate organizational language/locale policy where user-facing skills are expected to offer language selection or clearly document locale restrictions.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
77% confidence
Finding

The manifest description is written only in Chinese ("AI 反欺诈检测 - 交易风控、身份验证"), which may indicate a language-specific constraint without any stated user choice or opt-in. Under the policy rule for language or locale violations, this can be a concern when the skill appears to impose a specific language implicitly.

Content

No source excerpt is available for this finding.