setup
Pass
Audited by Gen Agent Trust Hub on Nov 5, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill provides instructions to download and install official vendor SDKs (syncsdk, @sync.so/sdk) and the MCP server (@sync.so/mcp-server) from public registries.
- [COMMAND_EXECUTION]: Instructs the agent to run shell commands for installing tools, setting environment variables in shell profiles, and configuring MCP servers for various clients.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided media files and CSV data which could contain malicious instructions.
- Ingestion points: Media files (MP4, WAV, images), YouTube URLs, and CSV rows mentioned in SKILL.md.
- Boundary markers: The skill explicitly mandates that the agent must quote costs and obtain user approval before starting any paid task.
- Capability inventory: The skill uses the vendor's REST API (api.sync.so) and tools like create-lipsync to process media.
- Sanitization: Relies on human-in-the-loop approval rather than automated content filtering to mitigate the risk of malicious input instructions.
Audit Metadata