Back to skill
Skillv1.0.0
ClawScan security
tmux-terminal · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
SuspiciousMar 6, 2026, 6:54 PM
- Verdict
- suspicious
- Confidence
- medium
- Model
- gpt-5-mini
- Summary
- The skill's instructions are coherent with a tmux-based terminal helper, but the declared metadata omits required tooling (tmux, and examples use cargo/rg), so the manifest and runtime instructions are inconsistent.
- Guidance
- This skill appears to be an instruction-only guide for using tmux, but the manifest omitted required tooling. Before installing or using it: (1) confirm tmux is installed and accessible on agents that will run it; update the skill metadata to declare tmux as a required binary; (2) note example commands also use cargo and rg — ensure those tools are expected/available if you plan to run those examples; (3) understand that following these instructions lets the agent start and control arbitrary local processes (it can run programs via tmux send-keys), so only enable autonomous invocation or run the skill in a trusted/sandboxed environment; (4) ensure sessions are cleaned up to avoid orphaned background processes. If you want to proceed, request the author fix the metadata to list required binaries (tmux, and optionally cargo/rg) so requirements match behavior.
Review Dimensions
- Purpose & Capability
- noteThe name/description (tmux-driven interactive terminal control) match the SKILL.md's guidance to create sessions, send keys, capture panes, and clean up. However the registry metadata lists no required binaries while the instructions explicitly require tmux and (in examples) cargo and rg; that omission is inconsistent.
- Instruction Scope
- noteSKILL.md tells the agent to run tmux commands that can launch and control arbitrary processes (e.g., cargo run). This is expected for a terminal automation skill, but the instructions reference additional tools (cargo, rg) and perform actions (starting/killing background processes, sending keystrokes) that operate at the system-shell level — these behaviors are broad and should be declared and reviewed.
- Install Mechanism
- okNo install spec and no code files — lowest install risk. Nothing is downloaded or written by the skill itself.
- Credentials
- okThe skill declares no environment variables or credentials and the instructions do not request secrets. Examples reference local build/runtime tools only; no external service credentials appear required.
- Persistence & Privilege
- okalways is false and the skill does not request system-wide persistence. The instructions do create background tmux sessions, which is within the skill's stated purpose; the user should be aware of leaving processes running if cleanup is missed.