Cli Builder

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:18 PM UTC
• Version:
  2.3.5

Overview

This is mostly a local CLI template generator, with one under-disclosed helper script that writes a small local command history but does not transmit data or perform destructive actions.

Install only if you want a local CLI template generator and are comfortable running bash scripts that call python3. Prefer scripts/cli-builder.sh for the documented generator features. Avoid passing secrets, tokens, or sensitive project names to scripts/script.sh because it can write command history under your local data directory.

SkillSpector

By NVIDIA

Vulnerability Patterns

• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• 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

Findings (3)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 87% confidence
• Finding: The skill advertises command generation features but does not declare permissions despite exhibiting capabilities to read environment data and read/write local files. Undeclared capabilities reduce transparency and bypass informed user consent, which is dangerous in agent settings because generated or supporting scripts may access sensitive local state unexpectedly.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 91% confidence
• Finding: The skill description claims a focused CLI scaffolding purpose, but the analyzed behavior indicates broader workflow automation plus persistent logging to a local history file and creation of an application data directory. This mismatch is risky because users may authorize the skill for code generation while it silently performs additional stateful actions and records command activity, which can expose project names, commands, paths, or other sensitive operational metadata.

Missing User Warnings

Medium

• Confidence: 93% confidence
• Finding: The script records command arguments verbatim to a persistent history file under the user's data directory without notice, consent, minimization, or redaction. Command arguments often contain secrets such as API keys, tokens, internal paths, or project names, so this creates a local sensitive-data exposure risk and can unintentionally retain confidential information.

VirusTotal

63/63 vendors flagged this skill as clean.