ClawHub
Back to skill

Security audit

building-twitter-industry-watchlist

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed Apify-based Twitter/X research workflow, with privacy and data-volume caveats but no evidence of hidden or destructive behavior.

Install this only if you are comfortable using Apify to collect and process Twitter/X research data. Keep APIFY_TOKEN private, avoid sensitive or regulated research targets unless approved, set maxItems deliberately, and avoid broad follower/following or retweeter harvesting unless it is necessary for your watchlist.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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 (6)

Description-Behavior Mismatch

Medium
Confidence
93% confidence
Finding
The manifest description states the skill returns a structured account list including engagement rate, topic focus, and influence score. In the file, the concrete executable steps are calls to `apidojo/twitter-user-scraper` and `apidojo/tweet-scraper`, while the scoring and account-type logic are only presented as prose and pseudocode without an implemented transformation pipeline, so the actual documented operation is scraper execution rather than guaranteed watchlist construction.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The skill requires an APIFY_TOKEN and relies on a third-party scraping service, but does not clearly disclose that user inputs and resulting data are transmitted to Apify. This creates privacy, compliance, and data-handling risk because users may unknowingly send sensitive market-research targets, account lists, or derived datasets to an external processor.

Context-Inappropriate Capability

Medium
Confidence
95% confidence
Finding
The exposed inputs allow collection well beyond the stated watchlist purpose, including follower/following graph extraction, retweeter harvesting, unavailable-user inclusion, and arbitrary JavaScript mapping. That broadens data collection scope and introduces unnecessary privacy and abuse risk, especially because the custom mapping hook could be used to transform or exfiltrate more data than users expect.

External Transmission

Medium
Category
Data Exfiltration
Content
**REST API fallback:**
```bash
curl -X POST \
  "https://api.apify.com/v2/acts/apidojo~tweet-scraper/runs?token=$APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"searchTerms": ["venture capital", "#vc", "VC trends 2026"], "maxItems": 500}'
Confidence
95% confidence
Finding
The REST example explicitly posts search terms and authenticates with an API token to an external Apify endpoint, confirming outbound transmission of user-supplied data to a third party. In this skill context that is expected functionality, but it is still security-relevant because it can leak sensitive research interests or operational data if users are not warned and controls are absent.

External Transmission

Medium
Category
Data Exfiltration
Content
**REST API fallback:**
```bash
curl -X POST \
  "https://api.apify.com/v2/acts/apidojo~tweet-scraper/runs?token=$APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"searchTerms": ["venture capital", "#vc", "VC trends 2026"], "maxItems": 500}'
```
Confidence
50% confidence
Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Missing User Warnings

Low
Confidence
90% confidence
Finding
Lines L063-L073 show commands that write results to local files, but the markdown provides no user-facing warning that the skill will persist collected account data on disk. Because this involves storing scraped data, the description should disclose the behavior so users can make an informed choice.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.