generate-test
Pass
Audited by Gen Agent Trust Hub on Apr 8, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill reads and processes external files (test-plan.md, backend-plan.md) to generate test code. This represents an indirect prompt injection surface where malicious instructions in these files could influence the behavior of the qa-specialist sub-agents.
- Ingestion points: **/test-plan.md, **/backend-plan.md, **/dev-plan.md
- Boundary markers: Absent; file content is used directly as context for sub-agents without delimiters or instructions to ignore embedded commands.
- Capability inventory: File system modification (tests/ directory), directory creation (mkdir), and sub-agent spawning (qa-specialist).
- Sanitization: No evidence of sanitization or filtering for the content of the ingested files.
- [COMMAND_EXECUTION]: The skill uses shell commands to initialize the testing environment and recommends standard development commands for the user to execute.
- Evidence: Executes 'mkdir -p' to create the test folder structure (tests/utils, tests/features, etc.). Suggests running 'npx vitest run' and 'npx supabase start' for test execution and environment setup.
- [EXTERNAL_DOWNLOADS]: Recommends the use of well-known developer tools and packages from official registries.
- Evidence: Mentions vitest and supabase CLI for test execution and local development environment management.
Audit Metadata