ClawHub
Back to skill

Security audit

ad-intelligence-skill

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed ad-intelligence guide for collecting public competitor ad data, but users should review scraping, third-party API, and dependency risks before using it.

Install/use this only if you are comfortable with public ad-library scraping and with sending competitor names, domains, page IDs, and API credentials to the named providers. Prefer official APIs where available, confirm platform terms and organizational policy, pin any dependencies before installing them, and run browser scraping in a restricted environment without unrelated secrets.

A.I.G

Vulnerability Patterns
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Findings (2)

T08 · Insecure Dependencies

Warning
Location
references/google.md:38
Finding

Unpinned Reverse-Engineered Google Ads Scraper Dependency

Content
View full analysis
Remediation
View remediation

T08 · Insecure Dependencies

Warning
Location
references/linkedin.md:40
Finding

Unpinned Playwright Package and Browser Artifact Installation

Content
View full analysis
Remediation
View remediation

SkillSpector

Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (12)

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · references/meta.md (reported line 128)May include surrounding context.

md
1. Facebook account (personal or business)
2. Verify identity at facebook.com/ID
3. Create a Meta Developer app at developers.facebook.com
4. Generate a User Access Token with `ads_read` permission
5. Token expires — refresh periodically (use long-lived tokens for automation)

### Key Limitation ⚠️

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill advertises activation for broad phrases like 'pull ad data', 'analyze competitor campaigns', and any request involving scraping or fetching public ad data across multiple platforms. That expansive trigger surface can cause the agent to invoke scraping-oriented behavior for loosely related marketing queries, increasing the chance of unintended data collection, policy-violating automation, or execution in contexts the user did not clearly request.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The document explicitly recommends scraping Google's Ads Transparency Center and using third-party wrappers, but it does not warn that queries, domains, advertiser IDs, and retrieved ad data may be transmitted to external providers or that scraping/reverse-engineering may violate platform terms or trigger operational/legal issues. In a skill meant for competitor intelligence, this omission can lead users or downstream agents to send potentially sensitive research targets to outside services without informed consent or policy review.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · references/linkedin.md (reported line 120)May include surrounding context.

md
Phase 1 alt: ScrapeCreators LinkedIn Ad Library API.
    100 free calls. signup at scrapecreators.com.
    """
    url = "https://api.scrapecreators.com/v1/linkedinadlibrary/search"
    headers = {"x-api-key": api_key}
    params = {"company": company_name}

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

This markdown file gives instructions for sending company/domain queries to external providers such as Adyntel, and similar sections later do the same for Apify and SociaVault, but it does not include any user-facing warning about sharing user-supplied data with third-party services. Under the markdown variant of SQP-2, externally transmitted data that could affect privacy should be disclosed in the skill description.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · references/linkedin.md (reported line 171)May include surrounding context.

md
Pass a company domain (e.g. 'salesforce.com') and get all their LinkedIn ads.
    Returns ads + continuation_token for pagination + total_count.
    """
    url = "https://api.adyntel.com/v1/linkedin/ads"
    headers = {"Authorization": f"Bearer {api_key}"}
    params = {"domain": domain}

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · references/linkedin.md (reported line 302)May include surrounding context.

md
def linkedin_ads_sociavault(company_name: str, api_key: str) -> list[dict]:
    """Phase 2c: SociaVault LinkedIn Ad Library API."""
    url = "https://api.sociavault.com/v1/scrape/linkedin-ad-library/search"
    headers = {"x-api-key": api_key}
    params = {"company": company_name}

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill explicitly instructs users to scrape Meta's ad library with headless browsers and unofficial tooling but provides no warning about platform terms, rate limits, privacy, or legal/compliance constraints. In an agent setting, this omission can normalize non-compliant collection practices and cause users to transmit search targets and retrieved ad/page data through unvetted scraping workflows without informed consent.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The document recommends unofficial and paid third-party APIs but does not disclose that brand queries, page identifiers, ad data, and API credentials will be sent to external vendors. That lack of transparency is risky because an agent may automatically forward sensitive research targets or secrets to external services the user did not intend to trust.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
93% confidence
Finding

This code sends user-supplied search terms or page IDs to ScrapeCreators, an external third-party service, without any warning, consent step, or data-handling boundary. In an agent workflow, that external transmission can expose competitive intelligence queries and associated metadata to a vendor outside the user's trust perimeter.

Content

Scanner excerpt · references/meta.md (reported line 109)May include surrounding context.

md
search_by: 'keyword' or 'page_id'
    """
    if search_by == "keyword":
        url = "https://api.scrapecreators.com/v1/facebookadlibrary/search"
        params = {"query": page_id_or_keyword}
    else:
        url = "https://api.scrapecreators.com/v1/facebookadlibrary/ads"

External Transmission

Medium
Category
Data Exfiltration
Confidence
95% confidence
Finding

The skill uses an x-api-key header with a third-party endpoint, meaning both the user's lookup target and a credential are involved in outbound transmission to an external service. Without warnings or guardrails, this raises the risk of inadvertent secret misuse, unapproved vendor access, and leakage of sensitive research activity.

Content

Scanner excerpt · references/meta.md (reported line 112)May include surrounding context.

md
url = "https://api.scrapecreators.com/v1/facebookadlibrary/search"
        params = {"query": page_id_or_keyword}
    else:
        url = "https://api.scrapecreators.com/v1/facebookadlibrary/ads"
        params = {"page_id": page_id_or_keyword}
    
    headers = {"x-api-key": api_key}

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
90% confidence
Finding

The embedded code sets Accept-Language to en-US,en;q=0.9, which imposes an English/US locale in the documented behavior. This is a natural-language locale constraint without opt-in or justification, so it matches the language/locale policy violation category.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.