ClawHub
Back to skill

Security audit

Appian Export

Security checks across malware telemetry and agentic risk

Overview

The skill does the advertised Appian export, but it handles credentials and output files more broadly than its user-facing documentation fully controls.

Review before installing. Use explicit environment variables rather than appian.json, run it from a directory where an appian-exports copy is acceptable, and prefer a version that validates the packageZip host and sanitizes downloaded filenames before use.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
91% confidence
Finding
The skill uses sensitive capabilities—environment variables for credentials and outbound network access to the Appian API—but does not explicitly declare permissions. That creates a transparency and policy-enforcement gap: operators may approve or run the skill without realizing it can access secrets and communicate externally, increasing the risk of unintended credential use or data export.

Description-Behavior Mismatch

Medium
Confidence
95% confidence
Finding
The code saves the exported ZIP to the declared storage path under the user's home directory and then silently copies it again into the current working directory. This expands the data exposure surface beyond what the manifest states, which matters because Appian exports can contain sensitive application logic, configuration, and package contents; the extra copy may land in a shared workspace, synced folder, or repository checkout unexpectedly.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

Detected: suspicious.env_credential_access

Environment variable access combined with network send.

Critical
Code
suspicious.env_credential_access
Location
scripts/index.js:33