Knowledge Base Index
architecture
Clinical AI Kit compiled knowledge on architecture.
| Article | Summary | Updated |
|---|---|---|
| Agent Architecture | A root orchestrator routes every request to one of three SequentialAgent pipelines. |
2026-07-04 |
| Diagram Atlas | The product now exposes the prepared architecture diagrams as an interactive atlas in the landing page, clinician dashboard, admin dashboard, and contextual workflow panels. | 2026-07-05 |
| Model Registry | capstone_agent/llm.py is the single place model selection happens. |
2026-07-04 |
| Module Dependency Graph | No circular dependencies. | 2026-07-04 |
| Module Reference | Every Python module currently in the repository, with its purpose (taken from module docstrings). | 2026-07-04 |
| System Overview | A multi-agent clinical AI platform built on Google ADK that processes medical imaging, answers patient questions with cited evidence, and runs natural-language database intelligence — all gated by clinician-in-the-loop review and privacy-aware security controls. | 2026-07-05 |
harness
Clinical AI Kit compiled knowledge on harness.
| Article | Summary | Updated |
|---|---|---|
| Claude Harness | Related: [[Development Workflow]] · [[Testing and Eval]]. | 2026-07-23 |
operations
Clinical AI Kit compiled knowledge on operations.
| Article | Summary | Updated |
|---|---|---|
| Clinical App | The clinician-facing product: a React frontend served by a FastAPI server that runs in deterministic demo mode by default or bridges live to the ADK agent backend. | 2026-07-05 |
| Deployment | Three supported targets are documented. | 2026-07-04 |
| MCP and A2A | Two interoperability surfaces let external clients and agents use Clinical AI Kit capabilities. | 2026-07-04 |
| Observability | Three pillars — Logs (diary), Traces (narrative), Metrics (timing) — plus clinical audit events. | 2026-07-04 |
| REST API and Developer Console | Clinical AI Kit exposes a versioned, secure REST API backend built on FastAPI and a React-based interactive Developer/API Console to assist developers in testing, auditing, and executing agent pipelines and MCP tools. | 2026-07-05 |
| Testing and Eval | pytest + pytest-asyncio for system contracts; ADK AgentEvaluator for content quality. |
2026-07-04 |
overview
Clinical AI Kit compiled knowledge on overview.
| Article | Summary | Updated |
|---|---|---|
| Course Concepts Map | The project covers all 10 course notebooks (Days 1a through 5b). | 2026-07-04 |
| Problem and Solution | Clinical work spans disconnected notes, session images, structured records, historical evidence, and population databases. | 2026-07-04 |
processes
Clinical AI Kit compiled knowledge on processes.
| Article | Summary | Updated |
|---|---|---|
| DB Intelligence Pipeline | SequentialAgent containing 6 specialist LLM agents translates natural language into safe SQL with validation and explicit approval preparation before execution, followed by chart-spec generation. |
2026-07-04 |
| Deployment Pipeline | Three targets fed by one hardened build. | 2026-07-04 |
| Development Workflow | Key facts:. | 2026-07-23 |
| End-to-End Request Flow | Every request crosses three security gates and one routing decision. | 2026-07-04 |
| Human-in-the-Loop Approval | The clinical review gate (Day 2b) — a LongRunningFunctionTool pauses the pipeline until a clinician decides. | 2026-07-04 |
| Image Extraction Pipeline | SequentialAgent containing 7 specialist LLM agents, with a nested LoopAgent, processes clinical images through quality assessment, OCR, AI vision analysis, structured field extraction, critic/refiner validation, and review-request preparation. |
2026-07-04 |
| Patient QA Pipeline | SequentialAgent containing 8 specialist LLM agents answers clinical questions with cited evidence from notes, images, and vector search — grounded in patient context. |
2026-07-04 |
security-memory
Clinical AI Kit compiled knowledge on security memory.
| Article | Summary | Updated |
|---|---|---|
| Memory Layers | Four-layer memory system with automatic PII/PHI redaction before anything persists. | 2026-07-04 |
| Security Layers | Defense in depth — three fixed callback layers plus clinical extensions. | 2026-07-04 |