Security audit

AI Data Analysis

Security checks for vulnerabilities and agentic risk

Overview

This skill is a simple local CSV analysis and cleaning tool with some documentation overclaims but no evidence of hidden, destructive, credential-seeking, or network behavior.

Before installing, treat this as a basic CSV-only local analysis utility, not the full-featured Excel/chart/PDF service described in the README. Run it only on data you intend to analyze, and note that it writes report or cleaned-data files into its local output directory.

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

MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
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 (5)

Natural-Language Policy Violations

Medium

Confidence: 92% confidence
Finding: The descriptive and instructional content is presented in Chinese only, with no indication that users may choose another language. Under SQP-3, forcing a specific language without opt-in is a natural-language policy concern unless the locale restriction is explicitly justified.

Missing User Warnings

Low

Confidence: 84% confidence
Finding: This markdown file documents writing analysis outputs such as `report.md` and `charts/`, and also advertises export features, but it does not include any warning that running the skill may create or overwrite local files. For markdown files, SQP-2 applies when behaviors affecting user data or system integrity are described without disclosure.

Intent-Code Divergence

Low

Confidence: 84% confidence
Finding: The function signature and CLI help indicate that callers can provide an `analysis_type` for `analyze`, implying different analysis behaviors may be selected. In practice, `analysis_type` is accepted at L13 and passed from the CLI at L76-L77, but never used; the code always generates the same overview and summary statistics report.

Natural-Language Policy Violations

Low

Confidence: 93% confidence
Finding: This code file contains natural-language strings in Chinese ("分析CSV文件" and "清洗数据") while the rest of the user-facing interface is in English. That creates an implicit language constraint without user opt-in or justification, which matches the language/locale policy violation criteria.

Natural-Language Policy Violations

Low

Confidence: 85% confidence
Finding: The package description includes Chinese text alongside English, which implies a specific language/localization choice without any indication that users can opt into their preferred language. Under the policy, language constraints should either offer a choice or be clearly justified as region-specific.

Static analysis

No suspicious patterns detected.