ClawHub
Back to skill

Security audit

extracting-tiktok-comments-for-research

Security checks across malware telemetry and agentic risk

Overview

This skill is a straightforward TikTok comment research workflow that uses Apify as disclosed, with some documentation and privacy-handling caveats.

Install only if you are comfortable sending TikTok video URLs and resulting public comment data to Apify. Use a least-privilege Apify token, prefer scoped direct video URLs over vague creator requests, and delete exported CSV/JSON datasets when they are no longer needed.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (7)

Description-Behavior Mismatch

Medium
Confidence
94% confidence
Finding
The manifest says the skill extracts TikTok comments from any video or creator, and Step 1 repeats that a creator handle can be used to pull comments from recent or viral videos. However, the actual documented actor inputs shown later only take post/video URL arrays (`startUrls`/`postURLs`) and provide no code or procedure for resolving a creator handle into videos before scraping comments.

Vague Triggers

Medium
Confidence
90% confidence
Finding
The trigger description is broad enough to activate the skill for generic research or audience-analysis requests that may not clearly imply scraping TikTok comments. Over-broad activation can cause unintended data collection and external transmission to Apify when the user did not explicitly request TikTok scraping, which is a consent and scope-control issue.

Intent-Code Divergence

Medium
Confidence
96% confidence
Finding
The Inputs table documents `startUrls`, `maxItems`, and `customMapFunction` as the skill parameters, but the actual run examples use different fields: `postURLs` and `maxCommentsPerPost`. This is not merely incomplete documentation; it presents conflicting parameter names for the same operation, so the stated interface diverges from the documented execution behavior.

External Transmission

Medium
Category
Data Exfiltration
Content
**REST API fallback:**
```bash
curl -X POST \
  "https://api.apify.com/v2/acts/apidojo~tiktok-comments-scraper/runs?token=$APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
Confidence
93% confidence
Finding
The skill explicitly sends user-supplied TikTok URLs and an APIFY_TOKEN-authenticated request to an external third-party service. External transmission is expected for this skill's purpose, but it still creates a real security/privacy boundary because user inputs and scraped data leave the local environment and are processed by Apify.

External Transmission

Medium
Category
Data Exfiltration
Content
**REST API fallback:**
```bash
curl -X POST \
  "https://api.apify.com/v2/acts/apidojo~tiktok-comments-scraper/runs?token=$APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "postURLs": [
Confidence
91% confidence
Finding
The hardcoded Apify API endpoint confirms dependence on an external network service for processing. While not inherently malicious, this expands the trust boundary and introduces risks around third-party handling of user inputs, metadata, and authentication material.

Missing User Warnings

Low
Confidence
88% confidence
Finding
The skill instructs saving scraped comments to local CSV/JSON files without any explicit warning that it will write files containing third-party user-generated content and usernames. This creates a quiet data-handling risk because users may not realize the agent is persisting scraped data locally, potentially retaining personal data longer than intended.

Natural-Language Policy Violations

Low
Confidence
93% confidence
Finding
The troubleshooting guidance suggests changing the search to English-language creators when comments are non-English. This steers use toward a specific language rather than offering the user a choice or framing it as an optional preference.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.