Security audit

Mimic Skill

Security checks for vulnerabilities and agentic risk

Overview

This skill is mostly a documentation/template helper, but it can guide broad extraction of sensitive clinical data and database changes without enough safety guardrails.

Install only if you have authorized MIMIC-IV access and will review generated queries before running them. Use read-only database credentials where possible, avoid broad all-patient extraction, store outputs securely, do not hardcode real passwords in shared code, and treat any CREATE INDEX suggestion as an administrator-only database operation.

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
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands

Findings (10)

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 89% confidence
Finding: The README promotes generation of ready-to-run SQL and Python for extracting ICU patient data from MIMIC-IV, but it does not clearly warn about handling protected health information, institutional approvals, minimum-necessary use, or output sanitization. In a healthcare data context, omission of privacy and governance guidance can encourage unsafe downstream use, overbroad extraction, or sharing of sensitive datasets even if the skill itself does not directly access the database.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 89% confidence
Finding: The manifest description is written in Chinese and specifies when to use the skill in Chinese, while the rest of the document is in English. This creates a language-specific activation/policy constraint without any user opt-in or explanation that the skill is intended only for Chinese-speaking users or a Chinese-only environment.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 92% confidence
Finding: The entire skill documentation, section headings, and usage guidance are written in Chinese, with no indication that another language is available or that Chinese is required for a justified regional/compliance reason. This creates a language/locale policy concern because the skill effectively mandates a specific language without user opt-in.
Content: No source excerpt is available for this finding.

Intent-Code Divergence

Medium

Category: Not specified by scanner
Confidence: 98% confidence
Finding: The Python example's docstring says it returns Charlson comorbidities, but the SQL only retrieves raw diagnosis rows for a single admission. This can mislead downstream agents or users into making clinical or analytical decisions based on uncomputed comorbidity data, creating integrity and misuse risks in a medical-data workflow.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 91% confidence
Finding: The document's headings, notes, and usage guidance are presented in Chinese, which effectively forces a specific language for users of the skill. Under the policy, language constraints should either be optional for the user or clearly documented and justified as region- or audience-specific.
Content: No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium

Category: Not specified by scanner
Confidence: 94% confidence
Finding: The reference includes CREATE INDEX ON chartevents (stay_id, itemid, charttime);, which is a schema-modifying SQL statement in a skill whose stated purpose is data extraction. Even though framed as a performance suggestion, executing DDL requires elevated privileges, changes the database state, and can cause operational impact such as locks, storage growth, or unauthorized modification of a shared clinical database environment.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The entire skill file is written in Chinese, including headings, notes, and usage guidance, with no indication that users may choose another language or locale. This can violate a language/locale policy when a skill imposes one language by default without explicit opt-in or documented regional justification.
Content: No source excerpt is available for this finding.

Missing User Warnings

Low

Category: Not specified by scanner
Confidence: 77% confidence
Finding: This markdown file contains a Python template that instructs users to provide database host, username, and password directly in code. Although the example is common, the surrounding documentation does not warn users against hardcoding credentials or suggest safer handling, which is relevant because markdown files should disclose behaviors affecting privacy or system integrity.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 95% confidence
Finding: 该 markdown 文件标题及全文内容均固定为中文表述，未说明这是面向特定中文用户群体的区域化文档，也未提供语言选择或英文替代。这符合“强制特定语言而无用户选择”的自然语言政策问题。
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 89% confidence
Finding: This markdown file presents all instructions, headings, and operational guidance exclusively in Chinese. Under the stated policy, forcing a specific language without user opt-in can be a natural-language policy violation unless the locale restriction is clearly justified.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.