neon-one
Pass
Audited by Gen Agent Trust Hub on Apr 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: Employs the
membraneCLI to perform API operations, search for actions, and manage connections. - [EXTERNAL_DOWNLOADS]: Requires the installation of the
@membranehq/clitool from the npm registry, which is the intended infrastructure for this skill. - [PROMPT_INJECTION]: Since the skill retrieves and processes external data from Neon One, it presents an indirect prompt injection surface (ingestion points:
membrane requestandaction runoutput). The skill does not implement specific boundary markers or content sanitization, and it has the capability to execute further shell commands based on agent logic (capability inventory:membraneCLI).
Audit Metadata