Trust Assessment
preview received a trust score of 80/100, placing it in the Mostly Trusted category. This skill has passed most security checks with only minor considerations noted.
SkillShield's automated analysis identified 3 findings: 0 critical, 1 high, 1 medium, and 1 low severity. Key findings include Unpinned or insecure dependency installation, Potential command injection via Docusaurus configuration/npm scripts, Broad filesystem and process interaction implied.
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 13, 2026 (commit 13146e6a). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.
Layer Breakdown
Behavioral Risk Signals
Security Findings3
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| HIGH | Unpinned or insecure dependency installation The skill explicitly states it "Installs dependencies from public npm registry". Without explicit version pinning (e.g., via `package-lock.json`) and using a command like `npm ci`, this process is vulnerable to supply chain attacks, where malicious packages or vulnerable versions could be pulled in, compromising the skill's execution environment. Ensure all Node.js dependencies are explicitly pinned to specific versions in `package.json` and `package-lock.json`. The skill's installation process should use `npm ci` to guarantee reproducible builds and prevent unexpected dependency updates from the public registry. | LLM | SKILL.md:30 | |
| MEDIUM | Potential command injection via Docusaurus configuration/npm scripts The skill describes starting a Docusaurus dev server and building a static site, which inherently involves executing `npm` commands (e.g., `npm start`, `npm run build`). If the `package.json` or Docusaurus configuration files (located in `.specweave/docs/` or generated dynamically) can be manipulated by an attacker, or if arguments to these commands are derived from untrusted input, arbitrary code execution could occur. The mention of "auto-fixes common issues" and dynamic configuration further increases this risk. Implement strict validation and sanitization for all inputs that influence file paths, configuration content, or command arguments used in `npm` scripts or Docusaurus commands. Ensure that `package.json` and Docusaurus configuration files are either immutable or generated from trusted, validated sources only. Avoid executing commands with user-controlled input directly. | LLM | SKILL.md:31 | |
| LOW | Broad filesystem and process interaction implied The skill's description and troubleshooting section imply broad filesystem write/read access (e.g., creating cache directories, `rm -rf` suggestions) and process interaction (e.g., starting servers, `lsof`, `kill -9` suggestions). While some access is necessary for its function, the potential for arbitrary file deletion or process termination if these actions are implemented without strict scope limitations poses a risk. Review the skill's actual implementation to ensure that filesystem operations (especially deletions) and process management are strictly confined to the necessary scope and do not allow arbitrary paths or process IDs. Implement robust input validation for any paths or identifiers derived from user input. | LLM | SKILL.md:103 |
Scan History
Embed Code
[](https://skillshield.io/report/63c7ebb1ef5754a6)
Powered by SkillShield