clip
Pass
Audited by Gen Agent Trust Hub on Apr 30, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill installs the
@membranehq/clipackage from the official NPM registry to facilitate platform interactions. - [COMMAND_EXECUTION]: The instructions guide the agent to execute shell commands using the
membraneCLI for managing connections, searching for actions, and performing data CRUD operations. - [DATA_EXFILTRATION]: The skill performs network operations to the vendor's infrastructure (Membrane) to process and store data. This is consistent with the skill's primary purpose of data management.
- [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection (Category 8).
- Ingestion points: Data returned from external actions via
membrane action run(e.g., Records data from Clip). - Boundary markers: None explicitly defined in the prompt templates.
- Capability inventory: Execution of CLI commands and creation of new server-side actions via
membrane action create. - Sanitization: Not explicitly documented; the skill relies on the underlying agent's standard handling of tool outputs.
Audit Metadata