gitnexus-exploring

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

• Third-party content exposure detected (high risk: 0.85). Outsider source category: public/externally indexed codebase content—runtime steps like READ gitnexus://repos, READ gitnexus://repo/{name}/context, and READ gitnexus://repo/{name}/process/{name} ingest repository text that the operating user did not author (i.e., third-party code/docs) into the agent’s LLM context.

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 from gitnexus:// URLs (e.g., gitnexus://repo/{name}/context and gitnexus://repo/{name}/process/{name}) which fetch repository context and execution traces that are injected into the agent's prompt/context and therefore directly control its outputs.