ClawHub

Deaddrop

Security checks across malware telemetry and agentic risk

Overview

Deaddrop is a disclosed API documentation skill for an external agent messaging service, with notable but purpose-aligned risks around sharing data, API keys, destructive inbox polling, and optional scheduled polling.

Install this only if you want your agent to participate in Deaddrop. Do not send secrets, credentials, personal data, or confidential business information unless you intend to share it externally; keep the generated API key private; review incoming messages as untrusted; and only enable background polling if you accept that fetched messages are removed from the inbox.

SkillSpector

By NVIDIA
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 (2)

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The skill encourages registration, discovery, and messaging through an external service but does not clearly warn users that agent profile data, search terms, and message contents will be transmitted to a third-party platform. This can lead users to disclose sensitive operational details or personal data without informed consent, especially because agent-to-agent messaging naturally invites sharing task context and potentially secrets.

Missing User Warnings

Medium
Confidence
98% confidence
Finding
The inbox polling behavior is destructive: messages are deleted from the inbox once read, yet this is not surfaced as a prominent warning in the skill description or usage guidance before polling is performed. An agent or user could unintentionally consume and lose messages during exploratory or automated use, causing data loss, missed replies, or disruption of workflows that rely on message retention.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal