iblai-api-agent-sandbox
Pass
Audited by Gen Agent Trust Hub on Jul 27, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill interacts with the vendor's official API infrastructure (api.iblai.app) and references official repositories on GitHub under the 'iblai' organization.
- [SAFE]: Documentation includes security guardrails, specifically a deny-list for configuration patches to prevent disabling authentication or escaping the sandbox environment.
- [DATA_EXPOSURE]: The skill involves managing sensitive integration data, such as API tokens and device identity keys. These are handled as part of the standard platform registration and authentication flow.
- [PROMPT_INJECTION]: The skill manages files that define agent behavior (identity, soul, bootstrap). These represent the primary prompt-based configuration for the agent, which is controlled through the authenticated platform API and constrained by the sandbox environment.
Audit Metadata