build-nitro-modules
Pass
Audited by Gen Agent Trust Hub on Mar 14, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill facilitates a potential indirect prompt injection vulnerability by instructing the agent to use unsanitized user input (library names) directly in shell commands such as
npx nitrogen@latest init <name>. \n - Ingestion points: User input for library name collected in
SKILL.mdandsetup-monorepo-init.md. \n - Boundary markers: None specified to delimit user input from instructions. \n
- Capability inventory: Execution of shell commands via
npx,bun,mv, andrmas described inSKILL.mdandsetup-monorepo-init.md. \n - Sanitization: The instructions do not include steps to validate or sanitize the user-provided string against shell metacharacters. \n- [EXTERNAL_DOWNLOADS]: The skill utilizes standard package managers to fetch and execute development tools and dependencies from well-known sources. \n
- Evidence: References to
npx nitrogen,npx @react-native-community/cli, andbun addthroughout the skill documentation for scaffolding and dependency management. \n- [COMMAND_EXECUTION]: The skill involves the execution of various shell commands to manage the filesystem and build processes. \n - Evidence: Documented use of
mv,rm,npx, andbunfor project setup, code generation, and running example applications.
Audit Metadata