Undeclared Tool Scope
Medium
- Category
- MCP Least Privilege
- Confidence
- 70% confidence
- Finding
Without declared permissions the skill's intent is opaque and cannot be validated.
- Content
Security audit
Crawler
Security checks for vulnerabilities and agentic risk
Overview
This is a documentation-only web crawling reference; it includes some anti-blocking advice to use cautiously, but I found no hidden execution, data collection, or persistence.
Install this as a reference skill only if you are comfortable with scraping guidance that includes anti-blocking techniques. Use it for lawful, permissioned crawling, prefer official APIs, follow site terms and robots.txt, and avoid using the proxy, fingerprint, CAPTCHA, or cookie-rotation advice to bypass a site's access controls.
Vulnerability Patterns
- 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
- Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
- Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
- Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
- MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
- Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
- Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (5)
Behavior Manipulation
Medium
- Category
- Prompt Injection
- Confidence
- 70% confidence
- Finding
Subtle instructions detected that may alter agent decision-making or introduce hidden biases.
- Content
sh - Always check ToS before scraping ## Ethical Scraping Guidelines 1. Check for API first (always prefer official API over scraping) 2. Read and follow robots.txt 3. Rate limit your requests (minimum 1s between requests per domain) 4. Identify your crawler (descriptive User-Agent with contact info)
Natural-Language Policy Violations
Medium
- Category
- Not specified by scanner
- Confidence
- 91% confidence
- Finding
The content explicitly recommends anti-detection measures such as rotating proxies, rotating headers, and varying TLS fingerprints to avoid blocking. In a web-scraping reference skill, this crosses from neutral education into operational guidance for evading site defenses, which can facilitate abusive or unauthorized scraping.
- Content
External Transmission
Medium
- Category
- Data Exfiltration
- Confidence
- 70% confidence
- Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
- Content
sh //div[contains(@class,"x")] Class contains //table/tr[position()>1] Skip header row ## curl with Cookie Jar curl -c cookies.txt -b cookies.txt -L "https://example.com/login" -d "user=x&pass=y" curl -b cookies.txt "https://example.com/data"
Natural-Language Policy Violations
Medium
- Category
- Not specified by scanner
- Confidence
- 95% confidence
- Finding
The FAQ gives step-by-step advice on how to avoid getting blocked, including rotating user agents, using residential proxies, matching browser fingerprints, and handling CAPTCHAs. That materially enables evasive scraping behavior and bypass of site-imposed defenses, making the skill more dangerous despite its reference-style framing.
- Content
Static analysis
No suspicious patterns detected.