Security audit

zoomeye-org-search

Security checks for vulnerabilities and agentic risk

Overview

The skill matches its ZoomEye search purpose, but it asks users to route a ZoomEye API key through the agent and installs an unpinned CLI package, so it needs review before use.

Before installing, prefer configuring ZoomEye yourself outside the chat, avoid pasting API keys into the agent, and rotate any key already shared this way. Install the CLI in a dedicated environment with a pinned trusted version where possible, and use `-save` only when you are comfortable storing search results locally.

A.I.G

Vulnerability Patterns

Insecure DependenciesIntroduces malicious components through unsafe dependency sources
Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
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

Findings (2)

T08 · Insecure Dependencies

Warning

Location: SKILL.md:50
Finding: Unpinned Third-Party Package Installation
Content: View full analysis
Remediation: View remediation

T09 · Insecure Skill Coding Practices

Error

Location: SKILL.md:58
Finding: ZoomEye API Key Exposed Through Chat and Command-Line Arguments
Content: View full analysis
" ``` It then validates the stored configuration with: ```bash zoomeye info ``` ### Technical Analysis A ZoomEye API key is a long-lived authentication credential. The Skill explicitly directs users to disclose this credential through the conversation and interpolates it into a shell command. This design can disclose the key through multiple channels: - Conversation history and AI service records - Agent traces and tool-call logs - Shell auditing or command logging - Process command-line inspection while the command runs - Diagnostic output, telemetry, or error reports - Persistent local ZoomEye configuration created by `zoomeye init` The command template also does not require validation of the key's character set before shell interpolation. Double-quote wrapping alone is not a general defense for arbitrary shell input, because shell expansions and quote termination can still become relevant if unexpected characters are accepted and the command is executed through a shell. ### Attack Path 1. The ZoomEye CLI reports that authentication is required. 2. Following the Skill instructions, the agent asks the user to send the API key in the conversation. 3. The user submits the credential, causing it to enter conversation records and agent processing logs. 4. The agent embeds the credential in the `zoomeye init -apikey` command. 5. The credential may additionally appear in tool traces, process arguments, command auditing, or error telemetry. 6. An operator, service component, local process, or other party with access to one of those records retrieves the API key ...[truncated 946 chars]
Remediation: View remediation

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (4)

Missing User Warnings

High

Category: Not specified by scanner
Confidence: 98% confidence
Finding: The skill explicitly instructs the user to send their ZoomEye API key to the agent for initialization, but provides no warning about credential sensitivity, retention, or safer alternatives. API keys are secrets that can be reused to access the user's account, consume quota, and expose account-linked activity, so collecting them through the agent materially increases credential-handling risk.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 90% confidence
Finding: The entire skill description, triggers, and operating guidance are presented only in Chinese, with no indication that users may choose another language or that the tool is intentionally limited to a Chinese-speaking context. Per the language/locale policy, forcing a specific language without user opt-in is a policy concern unless clearly justified.
Content: No source excerpt is available for this finding.

Ssd 3

Medium

Category: Not specified by scanner
Confidence: 97% confidence
Finding: The skill instructs the user to disclose their ZoomEye API key directly to the agent, unnecessarily routing a secret through the conversational system. Even if intended only for setup convenience, this increases the attack surface for accidental logging, leakage to third-party systems, or misuse of the credential.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 87% confidence
Finding: The skill documents use of -save to export search results to a local JSON file without warning that potentially sensitive reconnaissance data will be written to disk. This can lead to unintended persistence of asset inventories or exposure data in shared environments, backups, or endpoint monitoring systems.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.