relational-database-web-cloudbase
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides documentation and boilerplate code for the official
@cloudbase/js-sdk. The initialization patterns and database usage examples follow standard development practices for browser-side applications. - [EXTERNAL_DOWNLOADS]: The skill instructs the installation of the
@cloudbase/js-sdkpackage vianpm install. This is the official SDK from the vendor (tencentcloudbase) and is necessary for the skill's stated purpose. - [INDIRECT_PROMPT_INJECTION]: The skill facilitates reading data from a relational database, which introduces a potential surface for indirect prompt injection if retrieved content contains instructions intended to influence the agent's behavior.
- Ingestion points: Database query results retrieved via
db.from("posts").select("*")as described inSKILL.md. - Boundary markers: None explicitly defined in the provided snippets to delimit database content from agent instructions.
- Capability inventory: The SDK provides capabilities for data retrieval (
select) and modification (insert,update,delete) as documented inSKILL.md. - Sanitization: No specific sanitization or escaping mechanisms are provided in the boilerplate initialization code.
Audit Metadata