cursor-review-plugin-submission
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill uses a local Python utility (
scripts/cursor_plugin_submission_audit.py) to analyze plugin structures. This execution is limited to structural analysis and does not involve installing or activating the plugin components. - [INDIRECT_PROMPT_INJECTION]: The analysis script ingests untrusted data from the local plugin directory, including JSON manifests and Markdown files.
- Ingestion points: Reads
.cursor-plugin/plugin.json,README.md, and files within component directories (skills,rules,agents,commands) inscripts/cursor_plugin_submission_audit.py. - Boundary markers: The script uses a
_boundedfunction to ensure all file access remains within the provided plugin root, explicitly blocking absolute paths, directory traversal, and symlinks. - Capability inventory: The script is limited to read-only file access and data structure validation. It does not perform network requests, file writes, or execute code found within the plugin files.
- Sanitization: Employs standard library JSON parsing and regular expressions for metadata extraction, ensuring data is treated as structure rather than executable instructions.
- [EXTERNAL_DOWNLOADS]: The skill references source code and documentation from the official Cursor plugins repository on GitHub.
Audit Metadata