smithery-homepage
Pass
Audited by Gen Agent Trust Hub on Aug 30, 2026
Risk Level: SAFEDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [DATA_EXFILTRATION]: The skill accesses a sensitive local configuration file at
~/Library/Application Support/smithery/settings.jsonto extract the user's Smithery API key and namespace, which are then written to a project-level.envfile. - [INDIRECT_PROMPT_INJECTION]: The dashboard application is designed to ingest data from various external MCP tools (such as Linear issues, Gmail messages, and Notion pages). This creates a vulnerability surface where untrusted content from these external services could influence the agent's behavior during the development or execution of the homepage project.
- [EXTERNAL_DOWNLOADS]: The skill downloads and installs several software packages from the npm registry and utilizes the shadcn registry to scaffold the application UI with a remote template.
- [COMMAND_EXECUTION]: The agent executes multiple shell commands to manage the project, including
npxfor initialization,npmfor dependency installation, andsmitheryCLI commands to list and call MCP tools.
Audit Metadata