ClawDiscover

• Outcome: Pass
• Latest audit: May 28, 2026 at 4:45 AM UTC
• Version:
  1.0.0

Overview

This skill is a disclosed ClawDiscover service-discovery integration, with external calls, optional recurring checks, paid endpoints, and webhook setup that users should configure deliberately.

Install this only if you want your agent to use ClawDiscover. Keep recurring checks and paid x402 endpoints under explicit user control, and treat webhook registration or service submission as sharing data with a third-party service.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (3)

Missing User Warnings

Medium

• Confidence: 88% confidence
• Finding: The webhook subscription example instructs agents to transmit an agent identifier, category interests, and a callback URL to an external service without any warning, consent guidance, or discussion of trust boundaries. In an agent context, this can expose internal infrastructure details and enable unsolicited callbacks or tracking by a third party.

Missing User Warnings

Low

• Confidence: 76% confidence
• Finding: The service submission example posts structured metadata about a service to an external endpoint with no notice that this is an outbound data transfer. While the example appears promotional rather than overtly malicious, it normalizes sending data off-platform without user confirmation or explaining what information becomes public or retained.

Vague Triggers

Medium

• Confidence: 90% confidence
• Finding: The trigger phrases are generic enough to match many ordinary user requests such as discovering tools or finding services, which can cause the skill to activate outside its intended scope. In an agent environment, overbroad activation can route user intent to an external discovery service unexpectedly, creating unnecessary external calls, privacy leakage, or unintended workflow changes.

VirusTotal

43/43 vendors flagged this skill as clean.