Security audit

Full-Link Data Analysis

Security checks for vulnerabilities and agentic risk

Overview

This skill is a transparent data-analysis workflow, with the main caution that it may run Python and install analysis packages in the user's environment.

Install only if you are comfortable with an agent running Python over the data you provide. Do not allow automatic package installation unless you approve the package names and versions, preferably in an isolated virtual environment or sandbox. Avoid giving it sensitive raw datasets unless the environment and outputs are controlled.

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

Warning

Location: SKILL.md:89
Finding: Uncontrolled Installation of Unpinned Third-Party Dependencies
Content: View full analysis
Remediation: View remediation

SkillSpector

Vulnerability Patterns

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

Findings (5)

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 96% confidence
Finding: The trigger conditions are broad enough to activate on many ordinary analysis-related prompts, increasing the chance the skill runs in contexts where the user did not specifically request code-driven data processing. Overbroad activation raises the risk of unnecessary data access, Python execution, and downstream unsafe behaviors described later in the skill.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 89% confidence
Finding: The skill directs Python execution and possible package installation but provides no user-facing warning that code will run or that the system environment may change. This undermines informed consent and can surprise users with side effects, especially in environments where execution or installation has security and stability implications.
Content: No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium

Category: Not specified by scanner
Confidence: 92% confidence
Finding: The skill explicitly instructs the agent to install missing Python packages via pip during execution. Allowing package installation at runtime expands the execution surface, can modify the environment without user approval, and introduces supply-chain risk if package sources or names are influenced or not tightly controlled.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 83% confidence
Finding: Line L107 recommends jieba for Chinese word segmentation as part of the general text-analysis guidance, but the document does not indicate that this section is limited to Chinese-language text or provide an opt-in or alternative for other languages. Because this is natural-language guidance in a general methods library, the locale-specific recommendation can be read as a default language assumption rather than a documented regional constraint.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 83% confidence
Finding: The keyword list includes the Chinese-only term "突变" alongside English terms, but the document does not state that multilingual routing is optional or user-selected. This creates a mild language/locale policy concern because the skill implicitly embeds a specific locale without documenting choice or justification.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.