nika-authoring

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

• Third-party content exposure detected (medium risk: 0.65). The required workflow is to author a .nika.yaml and then run nika run <file>, which at runtime can execute invoke: tools like nika:fetch/traverse: (HTTP/site crawls) and/or exec: commands that may ingest arbitrary outsider-authored text, which the agent’s LLM context could then include via task prompts.