integration-ruby-on-rails
Pass
Audited by Gen Agent Trust Hub on Nov 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's workflow (specifically in
basic-integration-1.0-begin.md) requires the agent to read 10-15 files from the user's project to identify tracking opportunities. This creates a surface for indirect instructions embedded in user data to influence agent behavior. - Ingestion points: Project files analyzed by the agent during the setup phase (
basic-integration-1.0-begin.md). - Boundary markers: No specific boundary markers or instructions to ignore embedded prompts are provided for the file analysis phase.
- Capability inventory: The agent has capabilities to read, write, and modify files across the repository, as well as execute shell commands like
bundle installandrails generate. - Sanitization: No sanitization or validation of the content within the analyzed files is mentioned.
- [EXTERNAL_DOWNLOADS]: The skill references and downloads official PostHog components from trusted sources.
- Fetches the
@posthog/wizardvianpxfor automated installation as described inreferences/ruby-on-rails.md. - Loads the
posthog-jslibrary (array.js) from PostHog's asset delivery network in the application layout (references/EXAMPLE.md). - References official Ruby gems (
posthog-ruby,posthog-rails) from standard package registries. - [COMMAND_EXECUTION]: The skill instructs the agent and user to execute standard development commands for the Ruby on Rails framework.
- Includes usage of
bundle install,bin/rails db:migrate, andrails generate posthog:installto set up the environment and configuration.
Audit Metadata