Security audit

Tmux会话工具免费版

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed tmux helper that can read and type into terminal panes, so it is acceptable when that terminal access is intended.

Install only if you want an agent to inspect and interact with your tmux panes. Before sending input, confirm the session, window, and pane exactly; avoid using it on production shells or panes showing credentials, customer data, private prompts, or other sensitive output unless that access is intentional.

SkillSpector

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 (5)

Missing User Warnings

Medium

Confidence: 94% confidence
Finding: The skill provides a direct workflow for sending arbitrary keystrokes and pressing Enter in a targeted tmux pane, which can execute destructive commands if the target is wrong or influenced by untrusted input. Because tmux panes commonly host shells, REPLs, production sessions, or privileged tools, mis-targeting or unsafe target selection can lead to unintended command execution and system compromise.

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: Capturing recent pane contents and raw terminal buffers can expose credentials, API keys, tokens, proprietary code, command history, and private conversations displayed in the terminal. The skill presents this as routine functionality without a clear privacy warning or minimization guidance, increasing the likelihood of unnecessary disclosure.

Ssd 3

Medium

Confidence: 96% confidence
Finding: The skill explicitly instructs the agent to read recent tmux content and interpret user/assistant exchanges, creating a direct data-exfiltration path from terminal sessions into the agent's output. In real environments, those sessions may contain secrets, proprietary code, customer data, or operational details that should not be copied into summaries or responses.

Ssd 3

Medium

Confidence: 97% confidence
Finding: The buffer-dump feature increases exposure by extracting up to hundreds of lines of raw terminal history for debugging, which is likely to include sensitive material unrelated to the immediate task. Large raw dumps magnify both privacy risk and accidental leakage into logs, model context, or downstream systems.

Ssd 3

Medium

Confidence: 95% confidence
Finding: The best-practice section tells the agent to report user instructions and assistant replies from the tmux session, normalizing disclosure of conversational contents that may be sensitive. This increases the chance that private prompts, code, operational details, or embedded secrets are copied into summaries and shared more broadly than intended.

Static analysis

No suspicious patterns detected.