Security audit

Multi Search Engine Jarvis

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed search helper that sends user queries to listed search engines, with no local code execution, persistence, or credential access found.

Install only if you are comfortable with your search terms being sent to external search providers. Avoid entering secrets, credentials, personal data, or confidential investigation terms, and use safe-search or family-filter settings unless you intentionally need otherwise. Note that part of the documentation is Chinese-only and one referenced domestic guide file is missing from the package.

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
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (6)

Ae1

High

Category: analysis-evasion
Confidence: 100% confidence
Finding: Referenced artifact was not completely inspected
Content: Scanner excerpt · SKILL.md (reported line 105)May include surrounding context.
md
- `references/international-search.md` - International search guide

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 92% confidence
Finding: The natural-language content from the section headers through the usage examples is presented only in Chinese, which can amount to a language policy violation when no user opt-in or alternative language is offered. The file does not indicate that this skill is region-specific or that Chinese is required for compliance or domain reasons.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 94% confidence
Finding: The skill explicitly instructs direct web_fetch requests to third-party search engines using user-supplied queries, but it does not warn that those queries will be transmitted to external services. This can expose sensitive user input, internal investigation terms, or proprietary data to multiple search providers, some of which may log, profile, or retain requests.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 89% confidence
Finding: This manifest includes multiple China-region-specific engines and labels them with the fixed locale/region value "cn". Because the file provides no user opt-in, selection mechanism, or documented justification for enforcing region-specific providers, it creates a natural-language locale policy concern under the stated rules.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 93% confidence
Finding: The DuckDuckGo examples explicitly document turning off Safe Search (kp=-1) without any warning about adult, graphic, or otherwise unsafe content exposure. In a reusable agent skill, this can normalize unsafe defaults or lead downstream agents/users to issue broad searches that surface harmful material unexpectedly.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The Startpage parameter nj=0 is presented as a normal option to disable the family filter, again without warning about exposure to explicit or unsafe search results. Because this file is an instructional reference for agent behavior, it increases the chance that an agent or user will adopt the unsafe setting without understanding the consequences.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.