sc-e2e
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes shell commands using
npxto interact with testing frameworks. It triggers test suites, debuggers, and recording tools, which is necessary for its primary function as an E2E testing utility. - [INDIRECT_PROMPT_INJECTION]: The skill reads and processes external data, creating an attack surface for indirect prompt injection.
- Ingestion points: Test execution logs, source code components (via
ReadandGrep), and framework documentation (viamcp__pal__apilookup). - Boundary markers: No explicit delimiters or instructions are provided to the agent to ignore potentially malicious commands embedded in test failures or component code.
- Capability inventory: The skill possesses the ability to execute shell commands (
Bash) and write to the file system (Write). - Sanitization: No specific sanitization or validation steps for ingested file content or CLI output are defined.
Audit Metadata