v1.2.2

MOSS-TTS Voice

BenignClawScan verdict for this skill. Analyzed May 1, 2026, 6:21 AM.

Analysis

This skill appears to do what it says—generate and clone MOSS TTS audio—but it requires an API key and sends voice/audio data to MOSS servers.

GuidanceInstall only if you are comfortable sending selected text and voice samples to MOSS Studio. Use non-sensitive, consented audio, protect the API key, verify file paths before cloning, and clean up generated local files or stored voices when no longer needed.

Findings (4)

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.

Abnormal behavior control

Checks for instructions or behavior that redirect the agent, misuse tools, execute unexpected code, cascade across systems, exploit user trust, or continue outside the intended task.

Tool Misuse and Exploitation

SeverityLowConfidenceHighStatusNote

scripts/tts.py

parser.add_argument("--reference_audio", default=None, help="参考音频路径（实时克隆）"); with open(args.reference_audio, "rb") as f: payload["reference_audio"] = base64.b64encode(f.read()).decode("utf-8")

The reference-audio argument is a local file path that is read and sent in the API request; this is expected for cloning but depends on the user or agent selecting the intended audio file.

User impactChoosing the wrong path could upload an unintended local file to the MOSS API.

RecommendationPass only explicit, known audio-file paths and avoid using broad or sensitive directories as inputs.

Agentic Supply Chain Vulnerabilities

SeverityInfoConfidenceHighStatusNote

requirements.txt

requests>=2.28.0

The Python dependency is specified with a lower bound rather than an exact pinned version, so installs may resolve to different future versions.

User impactDependency behavior can vary across installations, though this is a common and limited dependency for an API client.

RecommendationInstall from trusted package sources, or pin dependencies in a local environment if reproducibility is important.

Permission boundary

Checks whether tool use, credentials, dependencies, identity, account access, or inter-agent boundaries are broader than the stated purpose.

Identity and Privilege Abuse

SeverityLowConfidenceHighStatusNote

references/troubleshooting.md

echo $MOSS_API_KEY # 应该显示: sk-xxx...

The skill relies on a MOSS API key and the troubleshooting guide suggests displaying it, which is useful locally but can expose the credential if terminal output is shared.

User impactAnyone who obtains the API key could use the associated MOSS account capabilities or quota.

RecommendationKeep MOSS_API_KEY out of screenshots, logs, chats, and repositories; rotate it if it is exposed.

Sensitive data protection

Checks for exposed credentials, poisoned memory or context, unclear communication boundaries, or sensitive data that could leave the user's control.

Insecure Inter-Agent Communication

SeverityMediumConfidenceHighStatusNote

SKILL.md

克隆音色：上传的音频会存储在 MOSS Studio 服务器; 实时克隆：每次请求都会上传音频数据到 MOSS API

The skill explicitly sends reference voice audio to an external provider and states cloned voice audio is stored by MOSS Studio.

User impactVoice recordings and text used for synthesis may leave the device and be retained by MOSS Studio.

RecommendationUse only recordings you have consent to clone, avoid sensitive speech, review MOSS Studio’s privacy policy, and delete unneeded stored voices.