jarvus-react
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of external data by instructing the agent to fetch library documentation using the context7 MCP server.
- Ingestion points: The
mcp__context7__query-docstool is used to retrieve content from external repositories (e.g., react-router, tailwindcss) inreferences/mcp-tools.md. - Boundary markers: There are no explicit instructions to delimit or treat the fetched documentation as untrusted data.
- Capability inventory: The skill context includes the ability to execute shell commands via
bunfor package installation and project scaffolding, and perform file writes across thesrc/directory. - Sanitization: No sanitization or validation mechanisms are defined for the content retrieved from external sources.
- [EXTERNAL_DOWNLOADS]: The skill references several external resources and tools required for the development environment.
- Downloads the
bunruntime manager plugin usingasdf plugin add buninSKILL.md. - Utilizes
bunx --bun shadcn@latestto fetch and execute the shadcn/ui CLI for component scaffolding inreferences/shadcn.md. - Fetches map styles and configuration from MapTiler's API in
references/maplibre.md. - [COMMAND_EXECUTION]: The skill relies on shell command execution for its core functionality.
- Instructs the agent to use
bun addandbun runfor dependency management and running development scripts. - Employs
asdf setandasdf installfor managing runtime versions.
Audit Metadata