yuki
Pass
Audited by Gen Agent Trust Hub on Apr 2, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill recommends installing the
@membranehq/cliglobal package via NPM. This tool is the official command-line interface for the Membrane platform (SKILL.md). - [COMMAND_EXECUTION]: The skill instructs the AI to execute various
membraneCLI commands to manage authentication, search for service connectors, and perform API operations (SKILL.md). - [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it retrieves data from the external Yuki API which is then provided to the agent.
- Ingestion points: Data retrieved from the Yuki API via
membrane action runandmembrane request(SKILL.md). - Boundary markers: Absent. There are no instructions for the agent to use delimiters or ignore instructions within the API response.
- Capability inventory: The agent can use the
membranetool to list actions, run actions, and make HTTP requests (SKILL.md). - Sanitization: Absent. No filtering or escaping of the remote API content is performed before it enters the agent's context.
Audit Metadata