cursor-memory-curator
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's primary function involves ingesting and processing untrusted data from existing Cursor rule files and memory-bank artifacts which could contain malicious instructions.
- Ingestion points: The skill reads repository context through
inventory-cursor-context.mjsandscan-cursor-context-risks.mjs, targeting files such as.cursor/rules/*.mdc,.cursorrules, andAGENTS.md. - Boundary markers: The skill does not utilize specific prompt delimiters or instructions to ignore embedded commands within the files it audits.
- Capability inventory: The skill possesses filesystem write and delete capabilities via
backup-cursor-context.mjsand the cleanup workflows defined inSKILL.md. - Sanitization:
scan-cursor-context-risks.mjsincludes aredact()function that utilizes regex patterns to mask sensitive data (tokens, passwords, emails) before the content is processed by the agent. - [COMMAND_EXECUTION]: The skill executes local Node.js scripts to perform repository auditing and file management. While the scripts are provided within the skill package and use static paths, they interact extensively with the local filesystem. Notably,
backup-cursor-context.mjsincludes sophisticated security checks, including symlink component detection, path snapshots, and file identity verification (fstat) to prevent directory traversal and TOCTOU attacks.
Audit Metadata