Doc refresh 1/2

Author: znorman-harris

extension/docs/github-copilot/getting-started/chat-basics.md

@@ -7,7 +7,7 @@ Use agent mode with Claude Sonnet 4.5 for the best experience with IDL and ENVI.
 **Requirements:**
 
 - Agent mode enabled (hexagon icon)
-- Claude Sonnet 4.5 recommended
+- Claude Sonnet recommended (4.5 at the time of writing this)
 - MCP tools visible in tool list
 
 ## Creating a New Chat

extension/docs/github-copilot/getting-started/envi.md

@@ -2,19 +2,32 @@
 
 ## Key Takeaways
 
-ENVI's GitHub Copilot integration uses specialized MCP tools to run processing workflows, understand datasets, and solve remote sensing problems using expert approaches.
+ENVI's GitHub Copilot integration uses specialized MCP tools to run processing workflows, understand datasets, and solve remote sensing problems.
 
 **Requirements:**
 
 - IDL 9.2 and ENVI 6.2 or newer
 - ENVI Agent license feature
 - Works alongside IDL Agent capabilities
 
-**Key Difference:** ENVI uses a guided "happy path" approach with dedicated tools, unlike IDL's open-ended programming focus.
+## Helpful Tips
+
+::: tip
+
+To guide the LLM to use ENVI, make sure to add "ENVI" to your requests. Here's some examples:
 
-## How It Works
+- Using ENVI can you do [awesome remote sensing workflow]
+- Can you [insert request] with ENVI?
 
-ENVI's integration centers on specialized MCP tools that enable GitHub Copilot to run ENVI processing, understand your datasets, and chain together workflows. This differs from the IDL integration—instead of open-ended programming, we guide the LLM through proven approaches using ENVI's established toolset.
+:::
+
+**Automatic Startup:** ENVI and IDL start automatically when needed—no manual intervention required.
+
+**Built-in Prompts:** Start your chat with "/envi" to access pre-configured prompts designed for common remote sensing problems.
+
+**Multi-File Data Sources:** For complex formats like Sentinel 2 or PlanetScope, provide the direct path to the root metadata file (e.g., "metadata.xml" for Sentinel 2 or "\_metadata.json" for Planet data). This ensures GitHub Copilot opens the correct dataset.
+
+**Understanding Data:** Ask GitHub Copilot to query or describe datasets when you need more information. This works for all data types, including complex multi-file formats.
 
 ## What You Can Do
 
@@ -34,7 +47,7 @@ View rasters, vectors, and raster series datasets directly in ENVI.
 
 Generate IDL routines and IDL Notebooks for repeatable workflows.
 
-## Example Prompts
+## Example Requests
 
 Guide GitHub Copilot to use ENVI by adding phrases like "Using ENVI", "With ENVI", or "Can ENVI" to your requests.
 
@@ -43,7 +56,7 @@ Here are some examples to get started:
 ::: code-group
 
 ```[Open Datasets]
-Can you open the ENVI file that lives "here" for me?
+Can you open the data in "here" for me in ENVI?
 ```
 
 ```[Query Datasets]
@@ -61,54 +74,3 @@ and display the results?
 ```
 
 :::
-
-### Working with ENVI's Tools
-
-**Built-in Prompts:** Start your chat with "/envi" to access pre-configured prompts designed for common remote sensing problems. These provide ready-to-use templates you can copy and customize.
-
-Here are additional examples for exploring ENVI's capabilities:
-
-::: code-group
-
-```[Tool Questions]
-Does ENVI have any tools for classification cleanup?
-```
-
-```[Planning #1]
-Can you help me come up with a change detection workflow using
-spectral indices with ENVI?
-
-I have imagery located in "this" folder that I want to use.
-
-I just want to plan a processing workflow, not run anything.
-```
-
-```[Planning #2]
-/envi Can you help me detect change between two images using
-machine learning?
-```
-
-```[Workflows]
-Do you know any change detection workflows using ENVI?
-```
-
-```[IDL Notebooks]
-/envi Help me come up with a data preparation workflow for the
-sample data "here".
-
-I don't want the processing to happen, instead please create
-an IDL Notebook that I can then customize based on
-other data sources.
-```
-
-:::
-
-## Helpful Tips
-
-**Automatic Startup:** ENVI and IDL start automatically when needed—no manual intervention required.
-
-**Built-in Prompts:** Start your chat with "/envi" to access pre-configured prompts designed for common remote sensing problems.
-
-**Multi-File Data Sources:** For complex formats like Sentinel 2 or PlanetScope, provide the direct path to the root metadata file (e.g., "metadata.xml" for Sentinel 2 or "\_metadata.json" for Planet data). This ensures GitHub Copilot opens the correct dataset.
-
-**Understanding Data:** Ask GitHub Copilot to query or describe datasets when you need more information. This works for all data types, including complex multi-file formats.

extension/docs/github-copilot/getting-started/idl.md

@@ -2,20 +2,26 @@
 
 ## Key Takeaways
 
-GitHub Copilot becomes an expert IDL programmer through specialized root-level instructions that automatically apply when you work with IDL code or notebooks.
+IDL's GitHub Copilot integration enables GitHub Copilot to become an expert IDL programmer through specialized root-level instructions. These instructions are automatically applied when you work with IDL code or IDL Notebooks.
 
 **Requirements:**
 
 - IDL 9.2 or newer
 - Instructions automatically load when IDL code or notebooks are open or attached
 
-**What to Look For:** "GitHub Copilot: IDL" message in your chat when instructions load
+**What to Look For:** "IDl Agent" message in your chat when instructions load
+
+## Helpful Tips
+
+**Automatic Startup:** IDL starts automatically when needed—no manual intervention required.
+
+**Creating IDL Notebooks:** Explicitly request an "IDL Notebook" to ensure you get the native VSCode implementation rather than a Python/Jupyter notebook. Native IDL Notebooks require no additional setup or configuration.
 
 ## How It Works
 
 GitHub Copilot is already an expert programmer across many languages. We've enhanced this expertise with specialized instructions that help it write better IDL code and understand your existing routines. These instructions apply automatically when you work with IDL files or notebooks, no manual setup required.
 
-## Example Prompts
+## Example Requests
 
 GitHub Copilot has access to many tools and can handle a wide range of IDL-related tasks or requests. Here are some examples to get you started:
 
@@ -46,12 +52,6 @@ that goes through the bounding boxes one time (so it is more linear for performa
 
 :::
 
-## Helpful Tips
-
-**Automatic Startup:** IDL starts automatically when needed—no manual intervention required.
-
-**Creating IDL Notebooks:** Explicitly request an "IDL Notebook" to ensure you get the native VSCode implementation rather than a Python/Jupyter notebook. Native IDL Notebooks require no additional setup or configuration.
-
 ## Attaching Context
 
 GitHub Copilot can incorporate various sources to better understand your code and provide more relevant answers. Click the paperclip icon in your chat to explore available options.

extension/docs/github-copilot/index.md

@@ -1,20 +1,16 @@
 # GitHub Copilot Integration
 
-IDL for VSCode integrates seamlessly with GitHub Copilot, giving you AI-powered assistance for IDL programming and ENVI workflows. This integration is completely optional and designed to enhance your productivity.
+IDL for VSCode integrates seamlessly with GitHub Copilot, giving you AI-powered assistants for IDL programming and ENVI workflows. This integration is completely optional for you to use and designed to enhance your productivity and help understand IDL and ENVI, not replace users.
 
 **What You Get:**
 
+IDL Agent and ENVI Agent which are powered by:
+
 - Native GitHub Copilot support through Model Context Protocol (MCP)
 - Smart IDL code generation with context-aware suggestions
 - Automated ENVI workflows that mirror expert problem-solving approaches
 - Automatic setup, no manual configuration required
 
-**Requirements:**
-
-- IDL 9.2 or newer
-- ENVI 6.2 or newer (for ENVI features)
-- ENVI Agent license feature (for ENVI capabilities)
-
 ## How It Works
 
 This integration focuses on two powerful capabilities:
@@ -25,7 +21,7 @@ This integration focuses on two powerful capabilities:
 
 ## Understanding Model Context Protocol
 
-The integration uses Model Context Protocol (MCP), a standardized interface that allows LLMs to run tools and interact with software. Think of MCP as a bridge between GitHub Copilot and IDL/ENVI.
+The integration uses Model Context Protocol (MCP), a standardized interface that allows LLMs to run tools and interact with software. Think of MCP as a bridge between GitHub Copilot and IDL/ENVI on your computer.
 
 ## Automatic MCP Server Setup
 
@@ -47,6 +43,6 @@ IDL for VSCode includes an MCP server that configures and starts automatically,
 **ENVI:**
 
 - ENVI 6.2 or newer
-- ENVI Agent license feature required
+- ENVI Agent license feature required for ENVI MCP tools
 
 The ENVI Agent license enables specialized tools optimized for remote sensing workflows, ensuring repeatable and reliable problem-solving approaches.