ClawHub

Mastodon Scout

Security checks across malware telemetry and agentic risk

Overview

Mastodon Scout is a transparent read-only Mastodon viewer, but users must keep its bearer token limited to read scope and only use trusted instance URLs.

Install only if you are comfortable giving the skill a Mastodon OAuth token. Create a dedicated token with only the read scope, set the instance to the exact Mastodon server that issued the token, and do not run it with untrusted --instance URLs.

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
85% confidence
Finding
The skill requires access to environment variables and makes outbound network requests, but those capabilities are not explicitly declared as permissions in the skill metadata. This creates a transparency and policy-enforcement gap: reviewers and runtimes may underestimate what the skill can access, increasing the risk of unintended secret exposure or unauthorized external communication.

Intent-Code Divergence

Medium
Confidence
96% confidence
Finding
The skill accepts a user-controlled --instance value and then sends the Authorization bearer token to that origin for every API request. In a supposedly read-only Mastodon client, this creates a credential exfiltration path: an attacker can point the client at a malicious server, capture the token, and then use it against the real Mastodon account within the token's granted scopes.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal