agentica-server
Pass
Audited by Gen Agent Trust Hub on Mar 23, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists of documentation and configuration instructions for a local development environment. It describes architectural components and provides troubleshooting steps for expected developer workflows.
- [COMMAND_EXECUTION]: Provides bash commands to initialize local services (Claude proxy and Agentica server) and perform health checks via curl on localhost. These commands target local scripts and binary installations within the project's workspace.
- [PROMPT_INJECTION]: Includes a directive formatted as 'CRITICAL: Use ACTUAL tools. Never DESCRIBE using tools.' which is intended as an anti-hallucination measure for the agent's tool-calling behavior. This instruction is presented as a configuration fix for development purposes rather than an attempt to bypass safety protocols.
Audit Metadata