OpenCode Game Builder

• Outcome: Review
• Latest audit: May 29, 2026 at 9:32 PM UTC
• Version:
  1.0.0

Overview

This is a legitimate OpenCode helper, but it can delegate broad coding work to a full-access subagent without clear safety boundaries.

Install only if you trust your local OpenCode setup. Use it in a dedicated project directory or sandbox, keep version-control checkpoints, avoid running parallel agents in the same workspace, and do not expose sensitive credentials or private sessions unless needed.

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: 95% confidence
• Finding: The skill advertises activation for broadly defined coding tasks, which can cause the agent to invoke a powerful subagent in situations where the user did not clearly request delegation. Because this skill can spawn an ACP agent with file-system and shell access, overly broad triggering increases the chance of unnecessary execution and expands the attack surface for prompt-injection or unintended destructive actions.

Missing User Warnings

Medium

• Confidence: 98% confidence
• Finding: The skill explicitly states that OpenCode has full file-system access and can execute shell commands, but it provides no safety boundaries, consent requirements, or limitations on sensitive paths and dangerous operations. In the context of an auto-invokable coding subagent, this makes accidental or induced data loss, secret exposure, system modification, or malicious command execution materially more likely.

VirusTotal

63/63 vendors flagged this skill as clean.