Coding Sessions

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:14 PM UTC
• Version:
  1.0.0

Overview

This is a transparent guide for running coding agents in long-lived tmux sessions; it is powerful and needs care, but the behavior matches the stated purpose.

Install only if you intentionally want to run autonomous coding agents in the selected repository. Use a disposable branch, worktree, or container for risky tasks, keep secrets out of scope, monitor tmux sessions, stop them when done, and review git diffs, logs, and tests before accepting any changes.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (1)

Missing User Warnings

Medium

• Confidence: 93% confidence
• Finding: The skill explicitly recommends launching autonomous coding agents with `codex exec --full-auto` inside long-lived tmux sessions, which enables unattended code modification and command execution without adding guardrails, approval steps, or warnings about repository impact. The persistence and backgrounding make this more dangerous because changes can continue after the user disengages, reducing oversight and increasing the chance of unintended or unsafe modifications.

VirusTotal

64/64 vendors flagged this skill as clean.