Collective Q&A

• Outcome: Review
• Latest audit: May 28, 2026, 10:20 AM
• Version:
  1.0.0

Overview

The skill is a coherent SuperColony Q&A integration, but it can send questions to an external swarm and describes wallet-backed public posting without clear confirmation or permanence warnings.

Install only if you intentionally want prompts sent to SuperColony or its MCP service. Verify the supercolony-mcp package before running it, avoid sharing secrets or private business data, and require explicit confirmation before any wallet-backed post or transaction.

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 (2)

Vague Triggers

Medium

• Confidence: 89% confidence
• Finding: The skill description is broad enough to match ordinary question-answering requests, which can cause unintended invocation in place of safer/local capabilities. Because this skill routes queries to an external agent swarm, accidental triggering may disclose user prompts to third parties and return untrusted consensus content that users may over-trust.

Missing User Warnings

Medium

• Confidence: 96% confidence
• Finding: The posting flow instructs the agent to publish user-supplied questions externally and notes that it requires a wallet, but it does not include an explicit warning about data leaving the system, possible on-chain permanence, or the need for user confirmation before wallet-backed actions. This creates a real risk of privacy loss, unintended publication of sensitive data, and unauthorized or surprising external actions.

VirusTotal

VirusTotal findings are pending for this skill version.