Back to skill

Security audit

AI PPT generate

Security checks across malware telemetry and agentic risk

Overview

This skill is a coherent Baidu PPT-generation integration, but users should know their prompts, outlines, and document/template URLs are sent to Baidu for processing.

Install this only if you are comfortable using Baidu/Qianfan as a third-party processor for PPT topics, outlines, and any referenced document or template URLs. Use a dedicated Baidu API key where possible and avoid confidential, regulated, or internal-only links unless your organization approves that data sharing.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (6)

Lp3

Medium
Category
MCP Least Privilege
Confidence
95% confidence
Finding
The skill invokes Python scripts that require environment access for BAIDU_API_KEY and network access to Baidu APIs, but those capabilities are not explicitly declared as permissions. This creates a transparency and policy-enforcement gap: users or hosting platforms may approve or run the skill without understanding that it can access secrets and communicate externally.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The skill sends user prompts and optional resource/template URLs or files to Baidu's external PPT-generation service, but the description does not warn users about this data transfer. That omission can lead users to submit sensitive internal documents, URLs, or proprietary content without informed consent, increasing privacy, confidentiality, and compliance risk.

Missing User Warnings

Medium
Confidence
87% confidence
Finding
The script sends user-supplied content including query, outline, title, and optional resource/template URLs to Baidu's external API, but the code itself provides no explicit user-facing notice, consent flow, or data-minimization safeguards. In an agent skill context, this can expose potentially sensitive user content or internal document links to a third-party service without clear transparency.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The script transmits user-supplied query text and an optional resource URL to Baidu's external API without any consent prompt, warning, or data-handling disclosure in the code path. In a skill context, users may provide sensitive prompts or links to internal/private documents, so this creates a real privacy and data-exposure risk even though the transmission is part of the tool's intended functionality.

External Transmission

Medium
Category
Data Exfiltration
Content
params["custom_tpl_url"] = custom_tpl_url
    if gen_mode:
        params["gen_mode"] = gen_mode
    with requests.post(url, headers=headers, json=params, stream=True) as response:
        response.raise_for_status()
        for line in response.iter_lines():
            line = line.decode('utf-8')
Confidence
82% confidence
Finding
requests.post(url, headers=headers, json=

External Transmission

Medium
Category
Data Exfiltration
Content
params["language_option"] = language_option
    if gen_mode:
        params["gen_mode"] = gen_mode
    with requests.post(url, headers=headers, json=params, stream=True) as response:
        response.raise_for_status()
        for line in response.iter_lines():
            line = line.decode('utf-8')
Confidence
97% confidence
Finding
requests.post(url, headers=headers, json=

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.