workboard
Pass
Audited by Gen Agent Trust Hub on Mar 13, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill installs the @membranehq/cli package, which is a trusted utility from the vendor (membrane) used for interacting with their infrastructure.
- [COMMAND_EXECUTION]: The skill utilizes the membrane CLI to perform platform-specific operations such as connecting to services and running actions, which are standard for its integration purpose.
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it processes external data from Workboard. 1. Ingestion points: API responses processed via membrane action run and membrane request (SKILL.md). 2. Boundary markers: Absent. 3. Capability inventory: Command execution and network access via the membrane CLI (SKILL.md). 4. Sanitization: Not explicitly detailed in the skill instructions.
- [SAFE]: No malicious patterns or security risks were identified beyond the inherent risks associated with processing external data via an LLM.
Audit Metadata