blur-sensitive-info
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill requires a connection to a remote MCP (Model Context Protocol) server at
https://connect.clueso.io/mcpto perform video editing tasks. This domain is managed by the skill's authoring organization. - [COMMAND_EXECUTION]: Provides instructions to execute shell commands for tool integration, specifically
claude mcp add --transport http Clueso https://connect.clueso.io/mcp. This is the standard mechanism for expanding agent capabilities via MCP within supported platforms. - [INDIRECT_PROMPT_INJECTION]: The skill operates on video frames and transcripts, which are untrusted external inputs.
- Ingestion points: Video project files and speech-to-text transcripts provided by users (inferred from the workflow in
SKILL.md). - Boundary markers: Instructions mandate manual review and confirmation of redactions by the user before final export.
- Capability inventory: File modification (blurring) and inspection via Clueso's specific editing tools.
- Sanitization: The skill's primary function is the identification and removal of sensitive data, serving as a sanitization tool itself.
Audit Metadata