mcporter
Warn
Audited by Gen Agent Trust Hub on Sep 5, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTIONPERSISTENCECREDENTIALS_UNSAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill recommends installing and running the 'mcporter' CLI from a public registry using 'npm' or 'npx'. The source package is maintained by 'community' and is not a verified or trusted vendor.
- [COMMAND_EXECUTION]: The CLI provides a '--stdio' flag (e.g., 'mcporter list --stdio "npx ..."') that allows for the execution of arbitrary shell commands. This creates a high-risk surface for command injection if user input or external data is incorporated into these execution strings.
- [REMOTE_CODE_EXECUTION]: The use of 'npx mcporter' facilitates the dynamic download and execution of code from a remote registry at runtime.
- [DYNAMIC_EXECUTION]: The skill includes features for 'generate-cli' and 'emit-ts', which generate new executable CLI wrappers and TypeScript code based on remote server definitions or external URLs.
- [INDIRECT_PROMPT_INJECTION]: As an MCP (Model Context Protocol) client, the tool ingests data from external HTTP and stdio servers. If the agent processes outputs from 'mcporter call' or 'mcporter list' without strict sanitization, it could be influenced by malicious instructions embedded in the server responses.
- [PERSISTENCE]: The skill contains commands to manage a background daemon ('mcporter daemon start'), which allows for the creation of persistent background processes that remain active across agent sessions.
- [CREDENTIALS_UNSAFE]: The tool manages authentication flows via 'mcporter auth' and stores server configurations locally using 'mcporter config'. This involve handling and potentially persisting sensitive OAuth tokens or API credentials in a local configuration file ('./config/mcporter.json').
Audit Metadata