mmx-cli
Pass
Audited by Gen Agent Trust Hub on Nov 10, 2026
Risk Level: SAFE
Full Analysis
- [DATA_EXPOSURE_AND_EXFILTRATION]: The skill manages MiniMax API keys and OAuth credentials, storing them in local configuration files (
~/.mmx/config.jsonand~/.mmx/credentials.json). The instructions explicitly direct the agent to avoid printing or repeating literal API keys in shell transcripts or logs, opting instead for environment variables or saved configurations. This represents standard and secure secret management for CLI-based tools. - [EXTERNAL_DOWNLOADS]: The skill guides the user to install the
mmx-clipackage from npm. This package is the primary tool for the skill's functionality and originates from the vendor's ecosystem. - [COMMAND_EXECUTION]: The skill makes extensive use of the
mmxcommand-line utility to perform media generation, transcription, and search tasks. It also includes instructions for building the tool from source (bun run build) within a development environment, which is appropriate for a developer-oriented skill. - [INDIRECT_PROMPT_INJECTION]: The skill defines several ingestion points for untrusted data, including web search results, audio files for transcription, and images for description. These inputs are passed as arguments to the CLI tool. The use of clear CLI flags serves as a boundary marker, and the processing is consistent with the skill's stated purpose of content generation and analysis.
Audit Metadata