Trust Assessment
sql-gen received a trust score of 81/100, placing it in the Mostly Trusted category. This skill has passed most security checks with only minor considerations noted.
SkillShield's automated analysis identified 2 findings: 0 critical, 1 high, 1 medium, and 0 low severity. Key findings include Potential Data Exfiltration via Schema Upload, Unpinned Dependency in Usage Example.
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 Findings2
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| HIGH | Potential Data Exfiltration via Schema Upload The skill describes a tool (`ai-sql`) that accepts a local file path for a database schema (`--schema ./schema.sql`) and explicitly states that an `OPENAI_API_KEY` is required. This strongly suggests that the content of the local schema file will be sent to OpenAI's API for processing. Sending local file content to an external third-party service without explicit user consent or clear data handling policies constitutes a data exfiltration risk. If the skill's underlying tool sends schema content to an external LLM, this should be clearly disclosed to the user, and consent obtained. Consider implementing local schema processing or anonymization if possible. Ensure sensitive information is not included in schema files or is properly redacted before transmission. | LLM | SKILL.md:25 | |
| MEDIUM | Unpinned Dependency in Usage Example The usage examples for `npx ai-sql` do not specify a version (e.g., `npx ai-sql@1.0.0`). This means `npx` will fetch the latest available version of the `ai-sql` package. This introduces a supply chain risk, as a malicious update to the `ai-sql` package could be automatically executed on the user's system without their explicit approval of a specific version. Update usage examples to include a pinned version for `npx` commands (e.g., `npx ai-sql@latest` or `npx ai-sql@1.2.3`) to allow users to explicitly choose the version they are running, mitigating risks from unexpected malicious updates. While `@latest` still fetches the newest, it makes the intent explicit. Pinning to a specific version is generally safer. | LLM | SKILL.md:15 |
Scan History
Embed Code
[](https://skillshield.io/report/37270141aebb4094)
Powered by SkillShield