Mimimax Voice Clone +TTS
Security checks across static analysis, malware telemetry, and agentic risk
Overview
The skill mostly matches its stated MiniMax voice-cloning purpose, but its write-back feature stores user-controlled names inside the skill document, which could persist unsafe instructions if not sanitized.
Review this skill before installing. It appears to use MiniMax for the advertised voice cloning/TTS workflow, but you should only send audio you have rights to use, provide a limited MiniMax API key, and watch the SKILL.md write-back behavior. Prefer simple display names without Markdown, backticks, or newlines, and inspect the mapping section after cloning.
Static analysis
No static analysis findings were reported for this release.
VirusTotal
VirusTotal findings are pending for this skill version.
Risk analysis
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
A malicious or accidental display name could alter the skill document and potentially influence future agent behavior when the skill is loaded.
User-controlled display names and voice IDs are written into SKILL.md, which is later read as skill context. The visible code does not show escaping of Markdown/backticks/newlines or strict character limits before persisting this content.
new_row = f"- `{display_name}`: `{voice_id}` (updated: {now})" ... skill_md_path.write_text(rebuilt, encoding="utf-8")Store mappings in a separate data file or escape/sanitize names before writing to SKILL.md. Require simple safe characters for display names, show the user a diff before write-back, and document how to remove bad mappings.
Voice recordings and text prompts leave the local machine and are handled by MiniMax according to that provider's policies.
The skill clearly discloses that selected voice audio and TTS text are sent to MiniMax. This is expected for the stated purpose, but it is still a sensitive external data flow.
The script calls MiniMax APIs over HTTPS ... Upload clone audio: `POST /v1/files/upload` ... Speech synthesis: `POST /v1/t2a_v2`.
Only use audio and text you are comfortable sending to MiniMax, avoid sensitive/private recordings unless appropriate, and ensure you have consent to clone the voice.
The skill can act under the MiniMax account associated with the supplied key.
The script uses MiniMax API credentials from the environment. That is expected for this provider integration, but the registry metadata lists no required environment variables.
os.getenv("MINIMAX_API_KEY") or os.getenv("MINIMAX_KEY") or os.getenv("MINIMAX_GROUP_API_KEY")Use a least-privileged MiniMax API key if available, keep it out of logs and shared shells, and update registry metadata to declare the credential requirement.
Future installs could use a different requests version than the one originally tested.
The dependency is purpose-aligned and common, but it is not pinned to an exact version, so installs may resolve to different package versions over time.
requests>=2.28.0
Pin dependencies or use a lockfile for reproducible installs, especially in shared or production environments.