Security audit

Mini Analysis Column

Security checks for vulnerabilities and agentic risk

Overview

This is a small CSV column-statistics skill that reads a user-specified file and computes max, min, and average values, with no hidden network, persistence, or credential behavior found.

Install in a virtual environment and consider pinning pandas if reproducibility matters. Only run it on CSV files you intend to analyze, and use --debug only when it is acceptable to print the first few rows of the file.

A.I.G

Vulnerability Patterns

Insecure DependenciesIntroduces malicious components through unsafe dependency sources
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

Findings (1)

T08 · Insecure Dependencies

Note

Location: SKILL.md:20
Finding: Unpinned Third-Party Dependency Installation
Content: View full analysis
Remediation: View remediation

SkillSpector

Vulnerability Patterns

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
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (5)

Tp2

High

Category: MCP Tool Poisoning
Confidence: 85% confidence
Finding: Mixing characters from multiple Unicode scripts in a single identifier is a common technique to create visually ambiguous tool names.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 87% confidence
Finding: The trigger '评分效果' lacks clear scope about file type, column analysis, or numeric-statistics behavior, so it may overlap with many unrelated evaluation or scoring requests. This ambiguity increases the chance of incorrect routing to this skill, which could expose file-processing capabilities or produce misleading analysis in the wrong context.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 93% confidence
Finding: The trigger '帮我看看csv文件的表现' is broad and conversational, making it likely to match ordinary user requests outside a narrowly intended workflow. Overly broad triggers can cause unintended skill invocation, leading the agent to process local files or perform analysis when the user did not explicitly choose this skill.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 96% confidence
Finding: The help text for --debug is written only in Chinese, and the script also prints Chinese status/output labels elsewhere, indicating a fixed-language user experience. This can violate language/locale policy when no user opt-in or documented regional limitation is provided.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 98% confidence
Finding: The script prints success/failure and result labels in Chinese only, with no mechanism for locale selection. A fixed locale in user-facing text can be a natural-language policy violation unless the skill explicitly documents and justifies that restriction.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.