Back to skill

Security audit

Yes Tool

Security checks for vulnerabilities and agentic risk

Overview

The skill is a simple automatic-response generator, but its documentation promotes blind confirmations for installs and includes a destructive deletion example without clear safety limits.

Review this skill carefully before installing. The code is minimal, but its examples encourage using automatic confirmations with installers, setup scripts, and deletion commands; only use it with fully understood, trusted, non-destructive commands and bounded counts where possible.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Missing User Warnings

Critical
Category
Not specified by scanner
Confidence
99% confidence
Finding

Including yes-tool | rm -rf dir/* as a default usage example is extremely unsafe because it associates blind automated confirmation with a destructive deletion command. Even if the pipeline is semantically questionable for rm, the documentation still models an unsafe pattern that could be copied, adapted to other commands, or interpreted by agents as acceptable practice for bypassing safeguards on irreversible operations.

Content

No source excerpt is available for this finding.

Missing User Warnings

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill description explicitly promotes using the tool to automatically confirm prompts for unattended execution, but it provides no warning that this can bypass important safety or consent checks in downstream commands. In an agent/tooling context, normalizing blind auto-confirmation increases the chance of destructive actions being approved without human review.

Content

No source excerpt is available for this finding.

Missing User Warnings

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

The quick-start examples encourage piping repeated confirmations into package installation and setup scripts without any safety disclosure. This is dangerous because it trains users or agents to suppress interactive safeguards and accept unknown prompts from software that may request risky changes, licenses, privilege elevation, or destructive actions.

Content

No source excerpt is available for this finding.

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
98% confidence
Finding

The documented use of rm -rf dir/* is a classic destructive command pattern and, in this skill, it is presented in conjunction with an automatic response generator. That combination materially increases the risk of mass data loss by encouraging parameter abuse and normalization of unattended destructive execution, especially if adapted to real paths or embedded in automation.

Content

Scanner excerpt · SKILL.md (reported line 40)May include surrounding context.

bash
# Default usage (outputs "y" forever until pipe closes)
yes-tool | rm -rf dir/*

# Custom response for unattended install
yes-tool "YESSIR" | ./configure