cursor-ide-weapon
Pass
Audited by Gen Agent Trust Hub on Jul 25, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists entirely of documentation, architectural guides, and development templates for the Cursor IDE environment.
- [EXTERNAL_DOWNLOADS]: The skill references standard package installations from the official npm registry, specifically
@cursor/sdkand@modelcontextprotocol/sdk. These dependencies are appropriate and expected given the skill's focus on programmatic IDE automation and tool integration. - [CREDENTIALS_SAFE]: Analysis of configuration templates (e.g.,
mcp-json-template.json) and guides (e.g.,03-mcp-integration.md) shows that the skill explicitly instructs users to use environment variable interpolation (e.g.,${env:MY_API_KEY}) rather than hardcoding sensitive credentials. Examples in the research section correctly identify hardcoded DSNs as security anti-patterns. - [PROMPT_INJECTION]: No prompt injection or safety bypass patterns were detected. The instructions focus on establishing development conventions and IDE personas via custom modes without attempting to override underlying model safety guardrails.
Audit Metadata