Doubao Seedream & Seedance API Skill

Security checks across malware telemetry and agentic risk

Overview

The skill appears purpose-aligned for using Volcengine media-generation APIs, but users should notice the API-key use, local history storage, and external setup instructions.

Before installing, confirm you trust any external setup commands, configure your Volcengine API key carefully, expect prompts/images/video requests to be sent to Volcengine, and periodically review or clear the local ~/.volcengine state if it may contain sensitive content.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

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.

#
ASI04: Agentic Supply Chain Vulnerabilities
Low
What this means

If the user follows the external setup path, they are trusting code outside the reviewed ClawHub artifact package.

Why it was flagged

The recommended setup asks the user to run an external repository's install script, while the provided ClawHub install spec does not include that script.

Skill content
git clone https://github.com/Lychee-AI-Team/seedream-skill.git
cd seedream-skill
./install.sh
Recommendation

Review the external repository and install script before running it, or prefer installing only the reviewed dependencies and files you intended to use.

#
ASI03: Identity and Privilege Abuse
Low
What this means

Using the skill can consume Volcengine account quota or incur API charges tied to the configured key.

Why it was flagged

The skill requires a Volcengine API key for authenticated provider calls; this is expected for the integration but is sensitive account authority.

Skill content
Volcengine API Key ... export ARK_API_KEY="your-api-key-here"
Recommendation

Use a dedicated or least-privilege API key if available, keep it out of shared files, and monitor provider usage.

#
ASI06: Memory and Context Poisoning
Low
What this means

Generated-content history may remain on the local machine and could be reused for later suggestions or visible to other local processes/users depending on file permissions.

Why it was flagged

The skill persists preferences, task state, and operation history locally, which may include prompts, task identifiers, and result metadata.

Skill content
self.state_dir = state_dir or Path.home() / ".volcengine"
self.state_file = self.state_dir / "state.json"
self.tasks_file = self.state_dir / "tasks.json"
self.history_file = self.state_dir / "history.json"
Recommendation

Avoid putting sensitive information in prompts, and clear or protect ~/.volcengine history files if privacy matters.