gitnexus-exploring

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.85). Steps 1–5 read from gitnexus://repos / gitnexus://repo/{name}/context / gitnexus_query / gitnexus_context / gitnexus://repo/{name}/process/{name}, which at runtime ingests codebase text from repositories the operating user did not author (outsider-authored source files and derived prose/trace data).

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill performs runtime READs of gitnexus:// resources (e.g., gitnexus://repo/{name}/context and gitnexus://repo/{name}/process/{name}), which fetch external repository/context content that is injected into the agent's reasoning context and therefore directly controls prompts and is required for the skill to operate.