cursor-keybindings
Pass
Audited by Gen Agent Trust Hub on Mar 11, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill definition in
SKILL.mdrequests broad shell access viaBash(cmd:*). This level of privilege is excessive for the stated purpose of managing editor keybindings and presents a risk for arbitrary command execution. - [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection when processing user requests to modify configuration files.
- Ingestion points: User prompts requesting changes to keyboard shortcuts are processed as instructions to modify
keybindings.json. - Boundary markers: No delimiters or 'ignore instructions' warnings are defined to isolate user input from the agent's logic.
- Capability inventory: The skill is granted file-writing and bash execution capabilities.
- Sanitization: No input validation or sanitization mechanisms are specified to prevent malicious user input from triggering unintended actions.
Audit Metadata