Android Sharesdk Integration

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed Android ShareSDK setup helper, but it handles app secrets and project edits that users should review carefully.

Install only if you intend to let an agent modify an Android project for ShareSDK. Use a git branch, review diffs before applying changes, do not paste live secrets into chat, keep generated credential files out of public repositories, and approve Gradle sync only for a trusted project.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Description-Behavior Mismatch

Medium
Confidence
94% confidence
Finding
The skill is presented as an interactive guide, but it also directs automated project modification and execution of `./gradlew --refresh-dependencies`. Running build tooling against a user-supplied path can change files, fetch remote dependencies, and execute repository-defined Gradle logic, which expands the trust boundary beyond simple guidance.

Context-Inappropriate Capability

Medium
Confidence
88% confidence
Finding
The skill includes third-party login sample code even though its stated purpose is social sharing integration. Adding login flows unnecessarily broadens permissions, data handling, and attack surface, and may prompt developers to enable authentication features they did not intend to ship.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The template explicitly asks the user to provide multiple platform app secrets and other credentials, but it does not warn against pasting those secrets into chat or provide a safer collection mechanism. In an agent workflow, users may reasonably disclose live production credentials to the assistant, creating a real secret-exposure risk that could lead to account takeover or abuse of third-party platform integrations.

Missing User Warnings

Medium
Confidence
87% confidence
Finding
The statement that the system will 'automatically complete integration configuration' after the user says they are done implies automated project modification without explaining what files will be changed, what content will be generated, or whether confirmation will be requested. In a code-assistant context, this can mislead users into approving opaque changes to build files, manifests, or SDK configuration, increasing the risk of unsafe or unexpected modifications.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal