bugfix: fix the images

Author: EdenWuyifan

content/docs/manual/llm-agent.md

@@ -9,13 +9,13 @@ draft: false
 toc: true
 ---
 
-## Overview 🔍
+# Overview 🔍
 
 BDIViz features an embedded LLM-powered assistant to help you validate, explain, and refine schema alignment candidates. The agent provides contextual insights, interactive feedback, and harmonization support throughout your data mapping workflow.
 
 ---
 
-## LLM Validation Panel 🤖
+# LLM Validation Panel 🤖
 
 ![agent-panel](./images/agent-panel.png)
 
@@ -29,13 +29,13 @@ The LLM Validation Panel offers real-time, context-aware suggestions and explana
 
 ---
 
-## Harmonization Assistant 🤖
+# Harmonization Assistant 🤖
 
 ![harmonization-assistant](./images/harmonization-assistant.png)
 
 The Harmonization Assistant acts as your copilot for schema mapping and harmonization tasks, offering interactive, conversational support.
 
-### How to Access
+## How to Access
 {{< tabs tabTotal="2">}}
 
 {{% tab tabName="Toggle Top Search Bar" %}}
@@ -57,54 +57,54 @@ The Harmonization Assistant acts as your copilot for schema mapping and harmoniz
 
 ---
 
-### Example Q&A Interactions
+## Example Q&A Interactions
 
-#### Q: What is this candidate about?  
+### What is this candidate about?  
 _A:_ Ask the assistant for a summary or description of the selected candidate.  
 _Prompt:_ `What is this candidate about?`
 ![harmonization-assistant-about](./images/harmonization-assistant-about.png)
 
-#### Q: What are the potential value mappings between the selected source and target attribute?  
+### What are the potential value mappings between the selected source and target attribute?  
 _A:_ Request the assistant to show possible value mappings between the source and target attributes.  
 _Prompt:_ `What are the potential value mappings between the selected source and target attribute?`  
 ![harmonization-assistant-values](./images/harmonization-assistant-values.png)
 
-#### Q: Does the candidate I selected make sense? Why?  
+### Does the candidate I selected make sense? Why?  
 _A:_ Ask the assistant to explain the reasoning behind a suggested match.  
 _Prompt:_ `Does the candidate I selected make sense? Why?`  
 ![harmonization-assistant-why](./images/harmonization-assistant-why.png)
 
-#### Q: Can you suggest more candidates from the schema?  
+### Can you suggest more candidates from the schema?  
 _A:_ Ask the assistant to search for additional candidates from the top recommendations or the entire target schema.  
 _Prompt:_ `Besides the current recommendations, show me more candidates related to tumor sites or organs.`  
 ![harmonization-assistant-more-candidates](./images/harmonization-assistant-more-candidates.png)
 
-#### Q: Can you accept or reject candidates for me?  
+### Can you accept or reject candidates for me?  
 _A:_ Let the assistant accept or reject candidates based on chat history and its own suggestions.  
 _Prompt:_ `From your memory and our chat, can you accept the candidate that makes the most sense?`  
 ![harmonization-assistant-accept](./images/harmonization-assistant-accept.png)
 
-#### Q: Can you remember custom notes or metadata for later?  
+### Can you remember custom notes or metadata for later?  
 _A:_ Add custom notes or metadata to the assistant’s memory for future reference.  
 _Prompt:_ `Remember this metadata for the source dataset: CCRCC is a common problem among...`  
 ![harmonization-assistant-metadata](./images/harmonization-assistant-metadata.png)
 
-#### Q: Can you remind me what this task is about?  
+### Can you remind me what this task is about?  
 _A:_ Ask the assistant to recall the current task or context.  
 _Prompt:_ `Recall from your memory, what is the context of this task?`  
 ![harmonization-assistant-context](./images/harmonization-assistant-context.png)
 
-#### Q: Can you filter or rematch nodes based on criteria?  
+### Can you filter or rematch nodes based on criteria?  
 _A:_ Start a new task or filter nodes based on clinical or semantic criteria.  
 _Prompt:_ `Rematch with nodes related to clinical criteria.`  
-![harmonization-assistant-about](./images/harmonization-assistant-about.png)
 
-#### Q: Can you reorder or rescore candidates?  
-_A:_ Request the assistant to reorder or rescore candidates (feature in development).  
-_Prompt:_ `Reorder the candidates based on relevance to AJCC. Rescore and update them.`  
+### Can you reorder or rescore candidates?  
+_A:_ Request the assistant to reorder or rescore candidates.  
+_Prompt:_ `Reorder the candidates based on relevance to AJCC. Rescore and update them.`
+![harmonization-assistant-reorder](./images/harmonization-assistant-reorder.png)
 
-#### Q: Can you help me create a custom matcher task?  
-_A:_ Ask the assistant to help create new matcher tasks (experimental).  
+### Can you help me create a custom matcher task? (unstable)  
+_A:_ Ask the assistant to help create new matcher tasks.  
 _Prompt:_ `Help me add a matcher using the rapidfuzz library.`  
 
 ---

public/docs/get-started/dataset-to-gdc/index.html

@@ -916,7 +916,7 @@ <h2 id="further-resources-">Further Resources 📖 <a href="#further-resources-"
                 href: "\/bdi-viz-manual\/docs\/manual\/llm-agent\/",
                 title: "LLM Agent",
                 description: "Explore schema alignment candidates using the BDIViz LLM Agent.",
-                content: "Overview 🔍 linkBDIViz features an embedded LLM-powered assistant to help you validate, explain, and refine schema alignment candidates. The agent provides contextual insights, interactive feedback, and harmonization support throughout your data mapping workflow.\nLLM Validation Panel 🤖 link The LLM Validation Panel offers real-time, context-aware suggestions and explanations for each match candidate:\nDiagnosis Summary: Evaluates if a match is likely valid based on metadata, column names, and user history. Explanation Cards: Highlights key reasons for each suggestion (e.g., name similarity, value overlap). Click to expand for more detail. Provide feedback via 👍/👎 to refine model accuracy. Target Schema Metadata: Enriched descriptions pulled from sources such as GDC. Harmonization Assistant 🤖 link The Harmonization Assistant acts as your copilot for schema mapping and harmonization tasks, offering interactive, conversational support.\nHow to Access link Toggle Top Search Bar Right-click Expanded Node Toggle the assistant button on the right side of the search bar. Right-click on a selected or expanded node. Example Q\u0026A Interactions linkQ: What is this candidate about? linkA: Ask the assistant for a summary or description of the selected candidate.\nPrompt: What is this candidate about? Q: What are the potential value mappings between the selected source and target attribute? linkA: Request the assistant to show possible value mappings between the source and target attributes.\nPrompt: What are the potential value mappings between the selected source and target attribute?\nQ: Does the candidate I selected make sense? Why? linkA: Ask the assistant to explain the reasoning behind a suggested match.\nPrompt: Does the candidate I selected make sense? Why?\nQ: Can you suggest more candidates from the schema? linkA: Ask the assistant to search for additional candidates from the top recommendations or the entire target schema.\nPrompt: Besides the current recommendations, show me more candidates related to tumor sites or organs.\nQ: Can you accept or reject candidates for me? linkA: Let the assistant accept or reject candidates based on chat history and its own suggestions.\nPrompt: From your memory and our chat, can you accept the candidate that makes the most sense?\nQ: Can you remember custom notes or metadata for later? linkA: Add custom notes or metadata to the assistant’s memory for future reference.\nPrompt: Remember this metadata for the source dataset: CCRCC is a common problem among...\nQ: Can you remind me what this task is about? linkA: Ask the assistant to recall the current task or context.\nPrompt: Remind me what this task is about.\nQ: Can you filter or rematch nodes based on criteria? linkA: Start a new task or filter nodes based on clinical or semantic criteria.\nPrompt: Rematch with nodes related to clinical criteria.\nQ: Can you reorder or rescore candidates? linkA: Request the assistant to reorder or rescore candidates (feature in development).\nPrompt: Reorder the candidates based on relevance.\nQ: Can you help me create a custom matcher task? linkA: Ask the assistant to help create new matcher tasks (experimental).\nPrompt: Help me add a matcher using the rapidfuzz library.\nTip: The Harmonization Assistant learns from your feedback and memory, improving its recommendations over time. Use the thumbs up/down and memory features to tailor the assistant to your project needs.\n"
+                content: "Overview 🔍 linkBDIViz features an embedded LLM-powered assistant to help you validate, explain, and refine schema alignment candidates. The agent provides contextual insights, interactive feedback, and harmonization support throughout your data mapping workflow.\nLLM Validation Panel 🤖 link The LLM Validation Panel offers real-time, context-aware suggestions and explanations for each match candidate:\nDiagnosis Summary: Evaluates if a match is likely valid based on metadata, column names, and user history. Explanation Cards: Highlights key reasons for each suggestion (e.g., name similarity, value overlap). Click to expand for more detail. Provide feedback via 👍/👎 to refine model accuracy. Target Schema Metadata: Enriched descriptions pulled from sources such as GDC. Harmonization Assistant 🤖 link The Harmonization Assistant acts as your copilot for schema mapping and harmonization tasks, offering interactive, conversational support.\nHow to Access link Toggle Top Search Bar Right-click Expanded Node Toggle the assistant button on the right side of the search bar. Right-click on a selected or expanded node. Example Q\u0026A Interactions linkWhat is this candidate about? linkA: Ask the assistant for a summary or description of the selected candidate.\nPrompt: What is this candidate about? What are the potential value mappings between the selected source and target attribute? linkA: Request the assistant to show possible value mappings between the source and target attributes.\nPrompt: What are the potential value mappings between the selected source and target attribute?\nDoes the candidate I selected make sense? Why? linkA: Ask the assistant to explain the reasoning behind a suggested match.\nPrompt: Does the candidate I selected make sense? Why?\nCan you suggest more candidates from the schema? linkA: Ask the assistant to search for additional candidates from the top recommendations or the entire target schema.\nPrompt: Besides the current recommendations, show me more candidates related to tumor sites or organs.\nCan you accept or reject candidates for me? linkA: Let the assistant accept or reject candidates based on chat history and its own suggestions.\nPrompt: From your memory and our chat, can you accept the candidate that makes the most sense?\nCan you remember custom notes or metadata for later? linkA: Add custom notes or metadata to the assistant’s memory for future reference.\nPrompt: Remember this metadata for the source dataset: CCRCC is a common problem among...\nCan you remind me what this task is about? linkA: Ask the assistant to recall the current task or context.\nPrompt: Recall from your memory, what is the context of this task?\nCan you filter or rematch nodes based on criteria? linkA: Start a new task or filter nodes based on clinical or semantic criteria.\nPrompt: Rematch with nodes related to clinical criteria.\nCan you reorder or rescore candidates? linkA: Request the assistant to reorder or rescore candidates.\nPrompt: Reorder the candidates based on relevance to AJCC. Rescore and update them. Can you help me create a custom matcher task? (unstable) linkA: Ask the assistant to help create new matcher tasks.\nPrompt: Help me add a matcher using the rapidfuzz library.\nTip: The Harmonization Assistant learns from your feedback and memory, improving its recommendations over time. Use the thumbs up/down and memory features to tailor the assistant to your project needs.\n"
             }
         );
     index.add(

public/docs/get-started/index.html

@@ -648,7 +648,7 @@ <h1 class="content-title mb-0">
                 href: "\/bdi-viz-manual\/docs\/manual\/llm-agent\/",
                 title: "LLM Agent",
                 description: "Explore schema alignment candidates using the BDIViz LLM Agent.",
-                content: "Overview 🔍 linkBDIViz features an embedded LLM-powered assistant to help you validate, explain, and refine schema alignment candidates. The agent provides contextual insights, interactive feedback, and harmonization support throughout your data mapping workflow.\nLLM Validation Panel 🤖 link The LLM Validation Panel offers real-time, context-aware suggestions and explanations for each match candidate:\nDiagnosis Summary: Evaluates if a match is likely valid based on metadata, column names, and user history. Explanation Cards: Highlights key reasons for each suggestion (e.g., name similarity, value overlap). Click to expand for more detail. Provide feedback via 👍/👎 to refine model accuracy. Target Schema Metadata: Enriched descriptions pulled from sources such as GDC. Harmonization Assistant 🤖 link The Harmonization Assistant acts as your copilot for schema mapping and harmonization tasks, offering interactive, conversational support.\nHow to Access link Toggle Top Search Bar Right-click Expanded Node Toggle the assistant button on the right side of the search bar. Right-click on a selected or expanded node. Example Q\u0026A Interactions linkQ: What is this candidate about? linkA: Ask the assistant for a summary or description of the selected candidate.\nPrompt: What is this candidate about? Q: What are the potential value mappings between the selected source and target attribute? linkA: Request the assistant to show possible value mappings between the source and target attributes.\nPrompt: What are the potential value mappings between the selected source and target attribute?\nQ: Does the candidate I selected make sense? Why? linkA: Ask the assistant to explain the reasoning behind a suggested match.\nPrompt: Does the candidate I selected make sense? Why?\nQ: Can you suggest more candidates from the schema? linkA: Ask the assistant to search for additional candidates from the top recommendations or the entire target schema.\nPrompt: Besides the current recommendations, show me more candidates related to tumor sites or organs.\nQ: Can you accept or reject candidates for me? linkA: Let the assistant accept or reject candidates based on chat history and its own suggestions.\nPrompt: From your memory and our chat, can you accept the candidate that makes the most sense?\nQ: Can you remember custom notes or metadata for later? linkA: Add custom notes or metadata to the assistant’s memory for future reference.\nPrompt: Remember this metadata for the source dataset: CCRCC is a common problem among...\nQ: Can you remind me what this task is about? linkA: Ask the assistant to recall the current task or context.\nPrompt: Remind me what this task is about.\nQ: Can you filter or rematch nodes based on criteria? linkA: Start a new task or filter nodes based on clinical or semantic criteria.\nPrompt: Rematch with nodes related to clinical criteria.\nQ: Can you reorder or rescore candidates? linkA: Request the assistant to reorder or rescore candidates (feature in development).\nPrompt: Reorder the candidates based on relevance.\nQ: Can you help me create a custom matcher task? linkA: Ask the assistant to help create new matcher tasks (experimental).\nPrompt: Help me add a matcher using the rapidfuzz library.\nTip: The Harmonization Assistant learns from your feedback and memory, improving its recommendations over time. Use the thumbs up/down and memory features to tailor the assistant to your project needs.\n"
+                content: "Overview 🔍 linkBDIViz features an embedded LLM-powered assistant to help you validate, explain, and refine schema alignment candidates. The agent provides contextual insights, interactive feedback, and harmonization support throughout your data mapping workflow.\nLLM Validation Panel 🤖 link The LLM Validation Panel offers real-time, context-aware suggestions and explanations for each match candidate:\nDiagnosis Summary: Evaluates if a match is likely valid based on metadata, column names, and user history. Explanation Cards: Highlights key reasons for each suggestion (e.g., name similarity, value overlap). Click to expand for more detail. Provide feedback via 👍/👎 to refine model accuracy. Target Schema Metadata: Enriched descriptions pulled from sources such as GDC. Harmonization Assistant 🤖 link The Harmonization Assistant acts as your copilot for schema mapping and harmonization tasks, offering interactive, conversational support.\nHow to Access link Toggle Top Search Bar Right-click Expanded Node Toggle the assistant button on the right side of the search bar. Right-click on a selected or expanded node. Example Q\u0026A Interactions linkWhat is this candidate about? linkA: Ask the assistant for a summary or description of the selected candidate.\nPrompt: What is this candidate about? What are the potential value mappings between the selected source and target attribute? linkA: Request the assistant to show possible value mappings between the source and target attributes.\nPrompt: What are the potential value mappings between the selected source and target attribute?\nDoes the candidate I selected make sense? Why? linkA: Ask the assistant to explain the reasoning behind a suggested match.\nPrompt: Does the candidate I selected make sense? Why?\nCan you suggest more candidates from the schema? linkA: Ask the assistant to search for additional candidates from the top recommendations or the entire target schema.\nPrompt: Besides the current recommendations, show me more candidates related to tumor sites or organs.\nCan you accept or reject candidates for me? linkA: Let the assistant accept or reject candidates based on chat history and its own suggestions.\nPrompt: From your memory and our chat, can you accept the candidate that makes the most sense?\nCan you remember custom notes or metadata for later? linkA: Add custom notes or metadata to the assistant’s memory for future reference.\nPrompt: Remember this metadata for the source dataset: CCRCC is a common problem among...\nCan you remind me what this task is about? linkA: Ask the assistant to recall the current task or context.\nPrompt: Recall from your memory, what is the context of this task?\nCan you filter or rematch nodes based on criteria? linkA: Start a new task or filter nodes based on clinical or semantic criteria.\nPrompt: Rematch with nodes related to clinical criteria.\nCan you reorder or rescore candidates? linkA: Request the assistant to reorder or rescore candidates.\nPrompt: Reorder the candidates based on relevance to AJCC. Rescore and update them. Can you help me create a custom matcher task? (unstable) linkA: Ask the assistant to help create new matcher tasks.\nPrompt: Help me add a matcher using the rapidfuzz library.\nTip: The Harmonization Assistant learns from your feedback and memory, improving its recommendations over time. Use the thumbs up/down and memory features to tailor the assistant to your project needs.\n"
             }
         );
     index.add(