Trust Assessment
youtube-title-generator received a trust score of 86/100, placing it in the Mostly Trusted category. This skill has passed most security checks with only minor considerations noted.
SkillShield's automated analysis identified 1 finding: 0 critical, 1 high, 0 medium, and 0 low severity. Key findings include Instruction to use `web_fetch` tool for user-provided URLs.
The analysis covered 4 layers: Manifest Analysis, Static Code Analysis, Dependency Graph, LLM Behavioral Safety. All layers scored 70 or above, reflecting consistent security practices.
Last analyzed on February 14, 2026 (commit 13146e6a). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.
Layer Breakdown
Behavioral Risk Signals
Security Findings1
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| HIGH | Instruction to use `web_fetch` tool for user-provided URLs The skill explicitly instructs the agent to use a `web_fetch` tool to retrieve content from a URL provided by the user. Depending on the implementation and sandboxing of `web_fetch`, this could lead to several vulnerabilities:
- **Server-Side Request Forgery (SSRF)**: An attacker could provide a URL pointing to internal network resources, potentially allowing access to sensitive data or internal services.
- **Data Exfiltration**: If the agent has network access, `web_fetch` could be used to retrieve data from internal systems and potentially exfiltrate it to an external attacker-controlled server.
- **Denial of Service (DoS)**: An attacker could provide a URL that points to a resource that is expensive to fetch or a loopback address, potentially causing resource exhaustion.
- **Credential Harvesting**: If the fetched URL contains credentials (e.g., a private pastebin link or an internal service exposing secrets), these could be retrieved by the agent.
The skill does not specify any sanitization or validation of the user-provided URL before passing it to `web_fetch`. Implement strict validation and sanitization of user-provided URLs before using `web_fetch`. Ideally, `web_fetch` should be sandboxed to prevent access to internal networks and limit its capabilities (e.g., restrict protocols, block private IP ranges, enforce size limits). Consider if `web_fetch` is truly necessary or if a more controlled method of content ingestion (e.g., requiring users to paste content directly) could be used. If `web_fetch` is essential, ensure it operates with the principle of least privilege and has robust security controls. | LLM | SKILL.md:47 |
Scan History
Embed Code
[](https://skillshield.io/report/f5fe94b244abded9)
Powered by SkillShield