update-cursor-settings
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest and process
settings.jsonfiles which are external to the skill and could potentially contain malicious configuration or be influenced by untrusted data sources. - Ingestion points: Accesses
~/Library/Application Support/Cursor/User/settings.json(macOS),~/.config/Cursor/User/settings.json(Linux), and%APPDATA%\Cursor\User\settings.json(Windows) as defined inSKILL.md. - Boundary markers: There are no instructions to use delimiters or specific boundary markers when the agent reads and interprets the configuration content.
- Capability inventory: The skill necessitates file system read and write permissions to update
settings.jsonand~/.cursor/cli-config.json. - Sanitization: The instructions recommend validating JSON syntax before writing changes back to the file, but do not provide guidance on sanitizing the semantic content of the settings to prevent malicious configuration changes.
Audit Metadata