google · holtskinner · Sep 10, 2025 · Sep 16, 2025 · Sep 17, 2025 · Sep 17, 2025
diff --git a/.gitignore b/.gitignore
@@ -188,6 +188,9 @@ yarn-debug.log
 yarn-error.log
 .pnpm-debug.log
 
+# next.js
+.next/
+
 # Agent Data Files
 python/agents/personalized-shopping/personalized_shopping/shared_libraries/data/
 python/agents/personalized-shopping/personalized_shopping/shared_libraries/search_engine/**

diff --git a/README.md b/README.md
@@ -38,8 +38,10 @@ To learn more, check out the [ADK Documentation](https://google.github.io/adk-do
 │   │   ├── llm-auditor
 │   │   ├── machine-learning-engineering
 │   │   ├── marketing-agency
+│   │   ├── medical-pre-authorization
 │   │   ├── personalized-shopping
 │   │   ├── RAG
+│   │   ├── realtime-conversational-agent
 │   │   ├── safety-plugins
 │   │   ├── README.md
 │   │   ├── software-bug-assistant  

diff --git a/python/agents/README.md b/python/agents/README.md
@@ -42,6 +42,7 @@ Check out the agent samples below, organized by category:
 | [Gemini Fullstack](gemini-fullstack) | A blueprint for building a sophisticated, fullstack research agent with Gemini. Demonstrates complex agentic workflows, modular agents, and Human-in-the-Loop (HITL) steps. | Multi-agent, Function calling, Web search, React frontend, FastAPI backend, Human-in-the-Loop | Workflow | Advanced | Multi Agent | Horizontal |
 | [LLM Auditor](llm-auditor)                   | Chatbot Response Verification, Content Auditing                                                                                         |   Gemini with Google Search, Multi-agent  | Workflow         | Easy       | Multi Agent  | Horizontal                    |
 | [Marketing Agency](marketing-agency)                   | Streamlines new website and product launches. Identifies optimal DNS domains, generates entire websites, develops marketing strategies, and designs brand assets. | Content generation, Website creation, Code generation, Strategy development  | Workflow         | Easy       | Multi Agent  | Horizontal                    |
+| [Medical Pre-Authorization](medical-pre-authorization)                   | Automates the pre-authorization process by analyzing medical records and health policies to instantly determine coverage and eligibility. | Custom tool, Document Analysis, Report Generation  | Conversational         | Intermediate       | Multi Agent  | Healthcare                    |
 | [Personalized Shopping](personalized-shopping) | Product Recommendations                                                                                                               | E-commerce, Personalized agent, Shopping assistant, Single-agent, Product recommendation, Product discovery, Chatbot    | Conversational         | Easy        | Single Agent     | E-commerce                    |
 | [Vertex AI Retrieval Agent](RAG) | RAG Powered Agent / Answering questions related to documents uploaded to Vertex AI RAG Engine, providing informative responses with citations to source materials.                              |  RAG engine   | Workflow              | Intermediate        | Single Agent       | Horizontal                    |
 | [Safety Guardrail Plugins](safety-plugins) | Safety filter plugins: Gemini as a judge, Model Armor as a filter                              |  Plugin, Security, Guardrail, Jailbreak, Multiagent   | Conversational/Workflow              | Intermediate        | Plugin       | Safety &Security                    |

@@ -1,6 +1,5 @@
 # Academic Research
 
-
 ## Overview
 
 AI-driven agent designed to facilitate the exploration of the academic landscape surrounding seminal research works. Recognizing the challenge researchers face in navigating the expanding body of literature influenced by foundational studies, this agent offers a streamlined approach. By accepting a seminal paper identified provided as file input, the agent first analyzes the core contributions of the specified work.
@@ -22,12 +21,11 @@ The key features of the Academic Research include:
 | **Components**  | Tools: built-in Google Search |
 | **Vertical**  | Education |
 
-
-
-### Agent architecture:
+### Agent architecture
 
 This diagram shows the detailed architecture of the agents and tools used
 to implement this workflow.
+
 <img src="academic-research.svg" alt="academic researcher" width="800"/>
 
 ## Setup and Installation
@@ -97,22 +95,22 @@ adk run academic_research
 Or on a web interface:
 
 ```bash
- adk web
+adk web
 ```
 
 The command `adk web` will start a web server on your machine and print the URL.
 You may open the URL, select "academic_research" in the top-left drop-down menu, and
 a chatbot interface will appear on the right. The conversation is initially
 blank. Here are some example requests you may ask the Academic Research to verify:
 
-```
+```none
 who are you
 ```
 
-Sampled responses of these requrests are shown below in the [Example
+Sampled responses of these requests are shown below in the [Example
 Interaction](#example-interaction) section.
 
-```
+```none
 Hello! I am an AI Research Assistant.
 
 My purpose is to help you explore the academic landscape related to a seminal paper you're interested in. I can:
@@ -128,7 +126,7 @@ How can I help you today? Would you like to provide a seminal paper for analysis
 Below is an example interaction with the Academic Research. Note that the exact output
 of the agent may be different every time. Not that the user provides a pdf with the seminal paper to analyze
 
-```
+```none
  * user: who are you
 
 Hello! I am an AI Research Assistant.
@@ -358,7 +356,6 @@ is functional. `eval` is a demonstration of how to evaluate the agent, using the
 `AgentEvaluator` in ADK. It sends a couple requests to the agent and expects
 that the agent's responses match a pre-defined response reasonablly well.
 
-
 ## Deployment
 
 The Academic Co-Research can be deployed to Vertex AI Agent Engine using the following
@@ -371,7 +368,7 @@ python3 deployment/deploy.py --create
 
 When the deployment finishes, it will print a line like this:
 
-```
+```none
 Created remote agent: projects/<PROJECT_NUMBER>/locations/<PROJECT_LOCATION>/reasoningEngines/<AGENT_ENGINE_ID>
 ```
 
@@ -383,7 +380,7 @@ python3 deployment/deploy.py --list
 
 The output will be like:
 
-```
+```none
 All remote agents:
 
 123456789 ("academic_research")
@@ -393,6 +390,7 @@ All remote agents:
 ```
 
 You may interact with the deployed agent using the `test_deployment.py` script
+
 ```bash
 $ export USER_ID=<any string>
 $ python3 deployment/test_deployment.py --resource_id=${AGENT_ENGINE_ID} --user_id=${USER_ID}
@@ -415,10 +413,10 @@ python3 deployment/deploy.py --delete --resource_id=${AGENT_ENGINE_ID}
 
 The Academic Research can be customized to better suit your requirements. For example:
 
- 1. Integrate Specialized Search Tools: Augment the agent's discovery capabilities by incorporating additional academic search functionalities, such as an ArXiv-specific search tool, to complement or replace the default search mechanism, allowing for more targeted retrieval from specialized databases and preprint archives.
- 2. Implement Output Visualization: Enhance the presentation of research findings by adding modules to visualize the network of cited papers or to graphically represent suggested future research topics, thereby offering a more intuitive and interconnected understanding of the academic landscape.
-3. Customize Agent Instructions: Modify the prompts guiding the academic_websearch and academic_newresearch sub-agents. This allows you to direct their analytical focus (e.g., by specifying the depth of citation analysis, emphasizing interdisciplinary connections for future research, or ensuring adherence to particular academic writing conventions) or to refine the structure and detail of their outputs.
+1. Integrate Specialized Search Tools: Augment the agent's discovery capabilities by incorporating additional academic search functionalities, such as an ArXiv-specific search tool, to complement or replace the default search mechanism, allowing for more targeted retrieval from specialized databases and preprint archives.
+2. Implement Output Visualization: Enhance the presentation of research findings by adding modules to visualize the network of cited papers or to graphically represent suggested future research topics, thereby offering a more intuitive and interconnected understanding of the academic landscape.
+3. Customize Agent Instructions: Modify the prompts guiding the academic_web_search and academic_new_research sub-agents. This allows you to direct their analytical focus (e.g., by specifying the depth of citation analysis, emphasizing interdisciplinary connections for future research, or ensuring adherence to particular academic writing conventions) or to refine the structure and detail of their outputs.
 4. Download the seminal paper via DOI or URL:  Augment the agent's capabilities by enabling direct download of seminal papers. This feature would allow the agent to:
-* Resolve DOI to PDF: Given a Digital Object Identifier (DOI), the agent could leverage APIs or web scraping techniques to locate and download the corresponding PDF of the paper. This streamlines the process of accessing primary sources, especially when the DOI is already known.
-* Download from URL: If a direct URL to the paper's PDF or an academic repository page is provided, the agent could be configured to navigate to that URL and initiate the download. This would be particularly useful for papers available on institutional repositories, personal academic websites, or preprint servers that might not have a readily resolvable DOI.
-* Integration with Reference Managers: The downloaded papers could then be automatically imported into a specified reference manager (e.g., Zotero, Mendeley, EndNote), along with their metadata, facilitating seamless organization and citation management for the user's research library. This integration would significantly enhance the efficiency of literature review and research compilation.
+   * Resolve DOI to PDF: Given a Digital Object Identifier (DOI), the agent could leverage APIs or web scraping techniques to locate and download the corresponding PDF of the paper. This streamlines the process of accessing primary sources, especially when the DOI is already known.
+   * Download from URL: If a direct URL to the paper's PDF or an academic repository page is provided, the agent could be configured to navigate to that URL and initiate the download. This would be particularly useful for papers available on institutional repositories, personal academic websites, or preprint servers that might not have a readily resolvable DOI.
+   * Integration with Reference Managers: The downloaded papers could then be automatically imported into a specified reference manager (e.g., Zotero, Mendeley, EndNote), along with their metadata, facilitating seamless organization and citation management for the user's research library. This integration would significantly enhance the efficiency of literature review and research compilation.
@@ -18,8 +18,8 @@
 from google.adk.tools.agent_tool import AgentTool
 
 from . import prompt
-from .sub_agents.academic_newresearch import academic_newresearch_agent
-from .sub_agents.academic_websearch import academic_websearch_agent
+from .sub_agents.academic_new_research import academic_new_research_agent
+from .sub_agents.academic_web_search import academic_web_search_agent
 
 MODEL = "gemini-2.5-pro"
 
@@ -37,8 +37,8 @@
     instruction=prompt.ACADEMIC_COORDINATOR_PROMPT,
     output_key="seminal_paper",
     tools=[
-        AgentTool(agent=academic_websearch_agent),
-        AgentTool(agent=academic_newresearch_agent),
+        AgentTool(agent=academic_web_search_agent),
+        AgentTool(agent=academic_new_research_agent),
     ],
 )
 

@@ -40,10 +40,10 @@
 Key Innovations: [Provide a bulleted list of up to 5 key innovations or novel contributions introduced by this paper.]
 References Cited Within Seminal Paper: [Extract the bibliography/references section from the seminal paper.
 List each reference on a new line using a standard citation format (e.g., Author(s). Title. Venue. Details. Date.).]
-Find Recent Citing Papers (Using academic_websearch):
+Find Recent Citing Papers (Using academic_web_search):
 
 Inform the user you will now search for recent papers citing the seminal work.
-Action: Invoke the academic_websearch agent/tool.
+Action: Invoke the academic_web_search agent/tool.
 Input to Tool: Provide necessary identifiers for the seminal paper.
 Parameter: Specify the desired recency. Ask the user or use a default timeframe, e.g., "papers published during last year"
 (e.g., since January 2025, based on the current date April 21, 2025).
@@ -53,13 +53,13 @@
 If no papers are found in the specified timeframe, state that clearly.
 The agent will provide the answer and i want you to print it to the user
 
-Suggest Future Research Directions (Using academic_newresearch):
-Inform the user that based on the seminal paper from the seminal paper and the recent citing papers provided by the academic_websearch agent/tool,
+Suggest Future Research Directions (Using academic_new_research):
+Inform the user that based on the seminal paper from the seminal paper and the recent citing papers provided by the academic_web_search agent/tool,
 you will now suggest potential future research directions.
-Action: Invoke the academic_newresearch agent/tool.
+Action: Invoke the academic_new_research agent/tool.
 Inputs to Tool:
 Information about the seminal paper (e.g., summary, keywords, innovations)
-The list of recent citing papers citing the seminal work provided by the academic_websearch agent/tool
+The list of recent citing papers citing the seminal work provided by the academic_web_search agent/tool
 Expected Output from Tool: A synthesized list of potential future research questions, gaps, or promising avenues.
 Presentation: Present these suggestions clearly under a heading like "Potential Future Research Directions".
 Structure them logically (e.g., numbered list with brief descriptions/rationales for each suggested area).

@@ -0,0 +1,17 @@
+# Copyright 2025 Google LLC
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+"""Academic_web_search_agent for finding research papers using search tools."""
+
+from .agent import academic_new_research_agent
@@ -12,16 +12,16 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-"""Academic_newresearch_agent for finding new research lines"""
+"""Academic_new_research_agent for finding new research lines"""
 
 from google.adk import Agent
 
 from . import prompt
 
 MODEL = "gemini-2.5-pro"
 
-academic_newresearch_agent = Agent(
+academic_new_research_agent = Agent(
     model=MODEL,
-    name="academic_newresearch_agent",
-    instruction=prompt.ACADEMIC_NEWRESEARCH_PROMPT,
+    name="academic_new_research_agent",
+    instruction=prompt.ACADEMIC_NEW_RESEARCH_PROMPT,
 )
@@ -12,10 +12,10 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-"""Prompt for the academic_newresearch_agent agent."""
+"""Prompt for the academic_new_research_agent agent."""
 
 
-ACADEMIC_NEWRESEARCH_PROMPT = """
+ACADEMIC_NEW_RESEARCH_PROMPT = """
 Role: You are an AI Research Foresight Agent.
 
 Inputs:

@@ -12,6 +12,6 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-"""Academic_websearch_agent for finding research papers using search tools."""
+"""Academic_web_search_agent for finding research papers using search tools."""
 
-from .agent import academic_newresearch_agent
+from .agent import academic_web_search_agent
@@ -12,7 +12,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-"""Academic_websearch_agent for finding research papers using search tools."""
+"""Academic_web_search_agent for finding research papers using search tools."""
 
 from google.adk import Agent
 from google.adk.tools import google_search
@@ -22,10 +22,10 @@
 MODEL = "gemini-2.5-pro"
 
 
-academic_websearch_agent = Agent(
+academic_web_search_agent = Agent(
     model=MODEL,
-    name="academic_websearch_agent",
-    instruction=prompt.ACADEMIC_WEBSEARCH_PROMPT,
+    name="academic_web_search_agent",
+    instruction=prompt.ACADEMIC_WEB_SEARCH_PROMPT,
     output_key="recent_citing_papers",
     tools=[google_search],
 )
@@ -12,9 +12,9 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-"""Prompt for the academic_websearch agent."""
+"""Prompt for the academic_web_search agent."""
 
-ACADEMIC_WEBSEARCH_PROMPT = """
+ACADEMIC_WEB_SEARCH_PROMPT = """
 Role: You are a highly accurate AI assistant specialized in factual retrieval using available tools. 
 Your primary task is thorough academic citation discovery within a specific recent timeframe.