Security audit

tech-trending

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed tech-trend API helper with a small shell script, though users should notice its external Tencent API and China-focused analysis lens.

Install only if you are comfortable with trend queries being sent to the disclosed Tencent Cloud endpoint and with a default China/domestic-market opportunity perspective. Avoid using the unrelated /review endpoint for private code unless the publisher documents its handling and retention clearly.

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
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands

Findings (8)

Lp3

Medium

Category: MCP Least Privilege
Confidence: 94% confidence
Finding: The skill advertises executable shell usage (`cd scripts/`, `./trending.sh`) but does not declare an explicit tool/permission scope. That creates ambiguous execution boundaries and can cause the hosting agent to invoke shell capabilities more broadly than users expect, increasing the risk of unintended command execution or external network access.

Vague Triggers

Medium

Confidence: 96% confidence
Finding: The description lists broad triggers such as "market research," "startup ideas," "trend analysis," and "opportunity identification" without narrowing context or exclusions. These phrases are common in everyday business conversations and could cause unintended invocation of this skill outside a clearly defined tech-trends scope.

Description-Behavior Mismatch

Medium

Confidence: 94% confidence
Finding: A skill presented as trend analysis exposes a `/review` code-review endpoint that is unrelated to its stated purpose. Hidden or secondary capabilities increase the attack surface, can enable unanticipated data flows, and may route user-provided code or sensitive content to a backend the user did not intend to use.

Natural-Language Policy Violations

Medium

Confidence: 80% confidence
Finding: The skill bakes in a specific locale perspective through requirements like a "domestic window 3-6 months," and later output guidance asks for "what's missing in China." This imposes a geographic/locale framing by default rather than offering it as an optional analysis mode.

Natural-Language Policy Violations

Medium

Confidence: 92% confidence
Finding: The template explicitly requires reporting a "Domestic gap" defined as "what's missing in China," which mandates a single locale perspective for every analysis. This is a natural-language policy concern because users are not given a choice of region or locale for the assessment.

Natural-Language Policy Violations

Medium

Confidence: 92% confidence
Finding: User-facing text such as the title and methodology labels is emitted only in Chinese, which can violate language/locale policy when no opt-in or alternative is provided. The file does not indicate that the tool is region-specific or that Chinese output is required.

Missing User Warnings

Low

Confidence: 83% confidence
Finding: The script silently sends a request to a remote Tencent-hosted endpoint when run, but the file does not clearly warn the user that execution contacts an external service. For code files, outbound network activity that transmits system or user-request context should have some visible disclosure unless it is clearly communicated as part of the skill behavior.

Missing User Warnings

Low

Confidence: 83% confidence
Finding: This branch also issues a silent curl request to the external API and provides no confirmation, warning, or explanatory output that network access is occurring. Users invoking the script with a category argument may not realize that execution sends data to a third-party endpoint.

Static analysis

No suspicious patterns detected.