TikHub API 工具(KK版)

• Outcome: Review
• Latest audit: May 29, 2026, 9:29 PM
• Version:
  1.1.0

Overview

The skill mostly does what it says, but its Python fallback can run unsafe background shell commands and implicitly uses a local TikHub API key.

Review before installing. Use a dedicated TikHub API key, monitor paid usage, avoid running the CPU/background transcription path on untrusted filenames or directories, and pin or review dependencies before using the Python fallback.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Behavioral AST: exec() Call, eval() Call, Dynamic Import

Findings (12)

subprocess module call

Medium

• Category: Dangerous Code Execution

Content

log_file = f"/tmp/whisper_{os.path.basename(audio_path)}.log"
        nohup_cmd = f"nohup {' '.join(cmd)} > {log_file} 2>&1 &"
        print(f"🚀 Whisper 后台转写启动，日志: {log_file}")
        subprocess.run(nohup_cmd, shell=True)
        print(f"📝 文字稿将保存到: {output_path}")
        print(f"⏱️  medium 模型 CPU 转写 1 分钟音频约需 1-2 分钟，请耐心等待")
        return output_path

• Confidence: 99% confidence
• Finding: subprocess.run(nohup_cmd, shell=True)

Context-Inappropriate Capability

Low

• Confidence: 83% confidence
• Finding: The code silently reads credentials from a fixed local .env path in the user's home directory without an explicit opt-in. In an agent skill context, implicit local secret discovery is risky because it expands the trust boundary and may cause the skill to use sensitive credentials the caller did not intend to expose.

Missing User Warnings

Medium

• Confidence: 93% confidence
• Finding: The balance-check example reads an API key from a local .env file and transmits it in an Authorization header to an external service, but the skill does not clearly warn the user that a local secret is being accessed and sent over the network. In an agent setting, normalizing secret access plus outbound transmission without explicit consent can lead to unintended credential disclosure or unsafe reuse patterns.

Missing User Warnings

Medium

• Confidence: 91% confidence
• Finding: The SDK example downloads remote content, writes files to /tmp, and invokes subprocesses (curl and ffmpeg) without an explicit warning that it changes the local system state. In agent-assisted workflows, undisclosed file writes and command execution are risky because they can surprise users, consume resources, and create opportunities for unsafe handling of untrusted media.

Vague Triggers

Medium

• Confidence: 88% confidence
• Finding: The manifest description advertises broad multi-platform data scraping capabilities but does not define any activation boundaries, user-intent constraints, or allowed use cases. In an agent skill ecosystem, vague scope increases the chance the skill is invoked in inappropriate contexts and can enable over-collection or misuse of scraping functionality across multiple services.

Missing User Warnings

Medium

• Confidence: 90% confidence
• Finding: Silently loading an API key from a local .env file reduces user awareness of credential access and makes secret use implicit. In agent environments, undisclosed local credential discovery can surprise users and enable unauthorized use of paid or sensitive accounts.

Missing User Warnings

Medium

• Confidence: 93% confidence
• Finding: The code launches a background process with nohup without requiring explicit acknowledgment, so work continues after the immediate call returns. In an agent setting, this reduces visibility and control, and when combined with shell execution it increases the chance of unnoticed abuse or persistence-like behavior.

Unpinned Dependencies

Low

• Category: Supply Chain

Content

requests
openai-whisper
mlx-whisper
ffmpeg

• Confidence: 97% confidence
• Finding: requests

Unpinned Dependencies

Low

• Category: Supply Chain

Content

requests
openai-whisper
mlx-whisper
ffmpeg

• Confidence: 91% confidence
• Finding: openai-whisper

Unpinned Dependencies

Low

• Category: Supply Chain

Content

requests
openai-whisper
mlx-whisper
ffmpeg

• Confidence: 91% confidence
• Finding: mlx-whisper

Unpinned Dependencies

Low

• Category: Supply Chain

Content

requests
openai-whisper
mlx-whisper
ffmpeg

• Confidence: 82% confidence
• Finding: ffmpeg

Known Vulnerable Dependency: requests — 10 advisory(ies): CVE-2014-1830 (Exposure of Sensitive Information to an Unauthorized Actor in Requests); CVE-2024-47081 (Requests vulnerable to .netrc credentials leak via malicious URLs); CVE-2024-35195 (Requests `Session` object does not verify requests after making first request wi) +7 more

High

• Category: Supply Chain
• Confidence: 99% confidence
• Finding: requests

VirusTotal

64/64 vendors flagged this skill as clean.