tapcart-blocks
Pass
Audited by Gen Agent Trust Hub on Mar 5, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the 'tapcart' CLI tool for creating and managing project assets. Commands such as 'tapcart block push' and 'tapcart component pull' are used to interact with the Tapcart environment.\n- [INDIRECT_PROMPT_INJECTION]: The agent reads local files ('code.jsx', 'config.json'), which is an injection surface.\n
- Ingestion points: Content from 'code.jsx' and 'config.json' is loaded into the agent's context during the push workflow.\n
- Boundary markers: None present.\n
- Capability inventory: The agent can execute CLI commands and host a local dev server.\n
- Sanitization: The instructions require the agent to verify the code content and obtain explicit user consent before proceeding with a push, serving as a human-in-the-loop safeguard.
Audit Metadata