frontegg-to-descope
Pass
Audited by Gen Agent Trust Hub on Nov 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSCREDENTIALS_UNSAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill possesses an indirect prompt injection attack surface.
- Ingestion points: Codebase analysis via grep patterns in 'references/frontegg-detection-patterns.md' and user input via the AskUserQuestion tool.
- Boundary markers: The skill requires human review of the generated 'MIGRATION-PLAN.md' and 'MIGRATION-STATE.md' before any code changes or project modifications are executed.
- Capability inventory: Capabilities include file system read/write, shell command execution for analysis, and network operations through the Descope MCP server.
- Sanitization: Codebase content used for analysis is not explicitly sanitized before interpolation into migration documents.
- [COMMAND_EXECUTION]: The skill utilizes 'grep' and 'find' commands to perform deep analysis of the user's codebase. These commands are configured to search for specific authentication patterns, environment variables, and dependency manifests.
- [EXTERNAL_DOWNLOADS]: The skill facilitates the migration process by recommending the installation of official Descope SDKs and interacting with vendor-hosted services, including the Descope MCP Server and Frontegg's regional management gateways.
- [CREDENTIALS_UNSAFE]: The migration process requires access to sensitive environment secrets (e.g., 'FRONTEGG_API_KEY', 'DESCOPE_MANAGEMENT_KEY'). The skill provides explicit guidance on the secure management of these keys, including warnings against client-side exposure.
Audit Metadata