smartbear
Pass
Audited by Gen Agent Trust Hub on Nov 11, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Downloads the
@membranehq/clipackage from the official npm registry to facilitate communication with the Membrane platform. - [COMMAND_EXECUTION]: Executes
membraneCLI commands to manage tenant login, connection establishment, and API request proxying. - [INDIRECT_PROMPT_INJECTION]: The skill processes data from SmartBear test cases and results which could theoretically contain malicious instructions.
- Ingestion points: Data returned from
membrane action runandmembrane requestcommands (SKILL.md). - Boundary markers: None present in the instructions to separate external data from agent instructions.
- Capability inventory: Network access via
membrane requestand command execution via the CLI (SKILL.md). - Sanitization: No explicit sanitization of the external API responses is mentioned.
- [SAFE]: No malicious behavior, obfuscation, or unauthorized data access patterns were identified. The skill correctly delegates credential management to a secure external handler.
Audit Metadata