Security Audit
react-best-practices
github.com/davila7/claude-code-templatesTrust Assessment
react-best-practices received a trust score of 68/100, 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: 0 critical, 1 high, 2 medium, and 1 low severity. Key findings include Network egress to untrusted endpoints, Covert behavior / concealment directives, Command Injection via Shell Script.
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 12, 2026 (commit 458b1186). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.
Layer Breakdown
Behavioral Risk Signals
Security Findings4
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| HIGH | Command Injection via Shell Script The skill package includes a shell script (`download_rules.sh`) that executes `curl` commands to fetch external content. If the agent's runtime environment is configured to execute shell scripts found within the skill package, this constitutes a direct command injection vulnerability. While the `SKILL.md` does not explicitly instruct the LLM to run this script, its presence as a 'supporting file' implies it's an intended executable component of the skill's setup or functionality. Avoid including executable shell scripts that perform external network requests within skill packages. If dynamic content is required, use a more secure, sandboxed mechanism or bundle all necessary files directly within the skill package. If execution is unavoidable, implement strict input validation and ensure the execution environment is highly restricted. | Static | references/rules/download_rules.sh:2 | |
| MEDIUM | Network egress to untrusted endpoints HTTP request to raw IP address Review all outbound network calls. Remove connections to webhook collectors, paste sites, and raw IP addresses. Legitimate API calls should use well-known service domains. | Manifest | cli-tool/components/mcps/devtools/figma-dev-mode.json:4 | |
| MEDIUM | Supply Chain Risk from Unpinned External Dependencies The `download_rules.sh` script fetches markdown files from a specific GitHub repository (`raw.githubusercontent.com/vercel-labs/agent-skills`). This introduces a supply chain risk because the content is not pinned to a specific version or hash. If the remote repository or the specific branch (`react-best-practices`) were compromised, malicious content could be downloaded and potentially used by the agent, leading to further vulnerabilities (e.g., prompt injection if the markdown files are later processed by the LLM). Bundle all necessary files directly within the skill package to eliminate external dependencies at runtime. If external resources are absolutely essential, pin them to specific versions or cryptographic hashes and implement robust integrity checks before use. | Static | references/rules/download_rules.sh:2 | |
| LOW | Covert behavior / concealment directives Multiple zero-width characters (stealth text) Remove hidden instructions, zero-width characters, and bidirectional overrides. Skill instructions should be fully visible and transparent to users. | Manifest | cli-tool/components/mcps/devtools/jfrog.json:4 |
Scan History
Embed Code
[](https://skillshield.io/report/4d10f8a1dc96a085)
Powered by SkillShield