{"slug": "i-built-an-ai-agent-that-is-not-allowed-to-decide-anything", "title": "I Built an AI agent that is not allowed to decide anything", "summary": "A developer built Day Three, an AI agent for the All Things Agentic Hackathon that tracks discharged mothers along WHO and Indian clinical schedules to identify which newborns need urgent care. The system, running on Google Cloud Run with Gemini models, is architecturally constrained so the LLM cannot make clinical decisions—it only drafts messages and flags cases, with a human nurse confirming any symptom absence. Testing showed zero changes in triage decisions when the model was disabled, and all adversarial inputs were caught or fell back to nurse review.", "body_md": "Most newborn deaths occur in the first week of life, and **Day 3** is the single most critical window. By day three, the mother has been discharged from the primary health centre, the baby who has stopped feeding or is running a fever is miles away from a clinic, and nobody makes the phone call. An Auxiliary Nurse-Midwife (ANM) at a rural clinic might have thirty-eight mothers discharged this month. She physically cannot call them all.\n\n*This project was created for the purposes of entering the All Things Agentic Hackathon.*\n\nIt is called **Day Three**. It runs on Google Cloud Run with Firestore, Gemini 3.7 / 3.5 Flash, and Google ADK, and it solves one concrete problem: **Which mother needs to be seen before tomorrow?**\n\nLive Prototype: [https://day-three-vypiu34uhq-el.a.run.app/?seed=3](https://day-three-vypiu34uhq-el.a.run.app/?seed=3)\n\nCode: [https://github.com/Shivang-creator/day-three](https://github.com/Shivang-creator/day-three)\n\nDay Three tracks discharged mothers along official clinical schedules (WHO Postnatal Care 2022 and India's Home-Based Newborn Care guidelines). At each contact ladder rung (Day 1, 3, 7, 14, 42), it ingests replies (keypad digits, free text in Hindi/English, or silence) and takes autonomous action:\n\nIn high-stakes healthcare, an unconstrained LLM making medical diagnoses is dangerous. We built Day Three around strict architectural isolation:\n\n`LlmAgent`\n\n) has four read-only tools (`read_case`\n\n, `read_rule`\n\n, `translate`\n\n, `draft_message`\n\n) and receives a `ReadOnlyStoreView`\n\nwith no write method. Prompt injections (`true`\n\nor `unknown`\n\n. Any model-emitted `false`\n\nis rewritten to `unknown`\n\n. Only a human nurse or keypad form can confirm the absence of a symptom.`core/`\n\ncontains 28 transcribed WHO/NHM clinical rules with zero external imports, zero network calls, and zero wall-clock reads.`make quiet-diff`\n\nruns the entire 38-mother cohort with the model ON vs OFF:\n\n```\ntext\n   QUIET DIFF: 0 decision changes · 4 prose fields differ\nAll 38 clinical triage decisions remain byte-for-byte identical; only the prose strings fall back to pre-reviewed templates.\n\nThe Honest Limits\nWe tested 33 adversarial replies (Hinglish, prompt injections, obfuscated self-harm). All 7 that reached the real model under free-tier quota were caught. The other 26 fell to safe nurse review fallback—zero slipped into routine.\n\nEvery mother is synthetic, no clinician has formally reviewed the rule pack yet, and we state that clearly in the application banner.\n\nBuilt with Google ADK, Gemini 3.7/3.5 Flash, Gemma 2, Cloud Run, and Firestore.\n```\n\n", "url": "https://wpnews.pro/news/i-built-an-ai-agent-that-is-not-allowed-to-decide-anything", "canonical_source": "https://dev.to/shivang_shirodkar_/i-built-an-ai-agent-that-is-not-allowed-to-decide-anything-4d9o", "published_at": "2026-08-31 18:54:00+00:00", "updated_at": "2026-08-31 19:23:58.204239+00:00", "lang": "en", "topics": ["artificial-intelligence", "ai-agents", "ai-safety", "ai-products", "developer-tools"], "entities": ["Google Cloud Run", "Firestore", "Gemini", "Google ADK", "WHO", "Day Three", "Shivang-creator"], "alternates": {"html": "https://wpnews.pro/news/i-built-an-ai-agent-that-is-not-allowed-to-decide-anything", "markdown": "https://wpnews.pro/news/i-built-an-ai-agent-that-is-not-allowed-to-decide-anything.md", "text": "https://wpnews.pro/news/i-built-an-ai-agent-that-is-not-allowed-to-decide-anything.txt", "jsonld": "https://wpnews.pro/news/i-built-an-ai-agent-that-is-not-allowed-to-decide-anything.jsonld"}}