youmind-deep-research
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the agent to install the
@youmind-ai/clipackage from the NPM registry. This package is a vendor-owned resource provided by the author to interface with the research service. - [COMMAND_EXECUTION]: The workflow relies on executing shell commands through the
youmindCLI to initialize research chats, poll for status, and retrieve results from the YouMind board. - [DYNAMIC_EXECUTION]: The skill uses
node -eto execute a short, hardcoded JavaScript snippet that parses and filters JSON data from standard input. This is used to extract specific research results from the API response messages. - [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect injection as it ingests arbitrary user-provided research topics and processes data retrieved from external web sources via the YouMind service.
- Ingestion points: Research questions and topics from user messages are interpolated into CLI command arguments in
SKILL.md. - Boundary markers: Not present; the skill treats user input as a direct search query.
- Capability inventory: The skill has access to shell execution (Bash), network operations (via the CLI), and Node.js execution.
- Sanitization: No explicit sanitization or escaping of the research topic is defined before it is passed to the CLI command.
Audit Metadata