collaborative-text-editor
Pass
Audited by Gen Agent Trust Hub on Jun 16, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill uses directive language such as 'IMPORTANT: Before doing anything, you MUST read BASE_SKILL.md' to ensure the agent follows established development patterns and error-handling policies. These are benign instructional markers intended to guide the AI agent's behavior within the platform's scope.
- [COMMAND_EXECUTION]: The documentation includes numerous examples of shell commands for project initialization, server management, and deployment, such as 'npm install rivetkit', 'npx @rivetkit/cli dev', and 'docker build'. These are standard development practices for the Rivet platform.
- [EXTERNAL_DOWNLOADS]: The skill references official RivetKit examples and templates hosted on GitHub (github.com/rivet-dev/rivet) and various NPM packages for framework integration. These sources are associated with the skill author and the core functionality of the framework.
- [SAFE]: The skill explicitly addresses security risks in its 'Security Checklist' section, advising users on authentication, input validation, and rate limiting. It identifies potential indirect prompt injection surfaces in unvalidated payloads and provides mitigation strategies, demonstrating a proactive security posture.
Audit Metadata