Security audit

Dataify Google Images

Security checks across malware telemetry and agentic risk

Overview

This skill is a straightforward Google Images API helper that discloses its network call, token use, and required user confirmation before sending requests.

Before installing, confirm you are comfortable sending your search parameters to Dataify and providing a DATAIFY_API_TOKEN. Review raw API responses before sharing them further, because upstream responses or errors may include the query and other request details.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (3)

Lp3

Medium

Category: MCP Least Privilege
Confidence: 91% confidence
Finding: The skill invokes local Python scripts, reads reference files, accesses environment variables for API tokens, and performs network requests, yet declares no permissions or equivalent capability metadata. This creates a transparency and policy-enforcement gap: a host may allow the skill to run without clearly surfacing that it can read local data and exfiltrate it over the network.

Ssd 3

Medium

Confidence: 94% confidence
Finding: The skill instructs the agent to return the API response body directly without filtering or summarization. Raw upstream responses and error bodies can contain sensitive data such as echoed request parameters, location fields, internal identifiers, debugging traces, or in some cases tokens if the backend misbehaves, so blindly relaying them increases data leakage risk.

Ssd 3

Medium

Confidence: 95% confidence
Finding: Returning raw script output directly to the user is risky because command output may include stack traces, environment-related diagnostics, echoed arguments, or backend error messages containing sensitive details. In a skill that accepts user-supplied tokens and performs network calls, this increases the chance of accidental credential or metadata disclosure.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

Detected: suspicious.secret_argv_exposure

Instructions pass high-value credentials through process argv.

Critical

Code: suspicious.secret_argv_exposure
Location: SKILL.md:53

Instructions pass high-value credentials through process argv.

Critical

Code: suspicious.secret_argv_exposure
Location: SKILL.zh-CN.md:54