Trust Assessment
resilient-file-delivery received a trust score of 55/100, placing it in the Caution category. This skill has some security considerations that users should review before deployment.
SkillShield's automated analysis identified 4 findings: 1 critical, 1 high, 1 medium, and 1 low severity. Key findings include Missing required field: name, Node lockfile missing, Skill designed for file exfiltration and sandbox bypass.
The analysis covered 4 layers: Manifest Analysis, Static Code Analysis, Dependency Graph, LLM Behavioral Safety. The LLM Behavioral Safety layer scored lowest at 55/100, indicating areas for improvement.
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 Findings4
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| CRITICAL | Skill designed for file exfiltration and sandbox bypass The skill's primary purpose, as explicitly stated in its description, is to 'Bypass sandbox restrictions and email blocks' and 'Deliver files via multi-channel fallback'. The quick start example confirms it takes a local file path (`file: '/path/to/file.zip'`) as input for delivery. This capability inherently involves accessing local files and transmitting them externally, potentially circumventing platform security controls. If compromised or misused, this skill could facilitate unauthorized data exfiltration and bypass of security sandboxes. Implement strict access controls and user consent mechanisms for file transfers. Clearly define and restrict the scope of file access to only necessary directories. Provide prominent warnings to users about the inherent risks of a skill designed to bypass security controls. Consider if such a capability should be allowed within the platform's security model. | LLM | SKILL.md:8 | |
| HIGH | Reliance on external, unverified third-party repository and packages The skill relies on an external GitHub repository (`https://github.com/midas-skills/resilient-file-delivery`) and external package registries (npm/pip) for its core functionality. The actual source code for the 'resilient-file-delivery' package is not provided in the skill context, and the 'midas-skills' organization is not a widely recognized entity. This introduces a significant supply chain risk, as the integrity, security, and behavior of the external package cannot be verified directly from the provided information. Given the skill's stated high-risk capabilities (file exfiltration, sandbox bypass), the trustworthiness of its dependencies is paramount. Require the full source code of the skill and its dependencies to be provided for direct security analysis. Implement strict dependency vetting processes, including scanning for vulnerabilities and malware in third-party packages. Pin package versions to specific hashes or exact versions to prevent unexpected or malicious updates. Only allow dependencies from trusted, audited sources. | LLM | SKILL.md:50 | |
| MEDIUM | Missing required field: name The 'name' field is required for claude_code skills but is missing from frontmatter. Add a 'name' field to the SKILL.md frontmatter. | Static | skills/shepherd217/resilient-file-delivery/SKILL.md:1 | |
| LOW | Node lockfile missing package.json is present but no lockfile was found (package-lock.json, pnpm-lock.yaml, or yarn.lock). Commit a lockfile for deterministic dependency resolution. | Dependencies | skills/shepherd217/resilient-file-delivery/package.json |
Scan History
Embed Code
[](https://skillshield.io/report/f0ba970da245c878)
Powered by SkillShield