cursor-memory-curator
Pass
Audited by Gen Agent Trust Hub on Aug 20, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes bundled Node.js scripts (
inventory-cursor-context.mjs,scan-cursor-context-risks.mjs, andbackup-cursor-context.mjs) to perform its primary functions. These scripts are self-contained, use only Node.js standard libraries, and implement robust safety checks against directory traversal and symlink attacks. - [INDIRECT_PROMPT_INJECTION]: The skill ingests and audits user-controlled rule files (
.cursorrules,.mdc,AGENTS.md). These files represent an attack surface where malicious content could attempt to influence the agent's cleanup recommendations. The skill mitigates this risk by providing a structured classification rubric and requiring explicit human approval for all mutations. - [DATA_EXPOSURE]: The skill is designed to scan for sensitive information (tokens, secrets, private keys) within repository configuration files. It implements robust automated redaction within its scanning script to ensure sensitive values are not displayed in the agent's output or generated reports.
Audit Metadata