New Guide Article : "Microbots : Introduction, Installation Guide and Creating Your First MicroBot" and structural changes in blog by shivashanmugam · Pull Request #132 · microsoft/microbots

shivashanmugam · 2026-04-17T11:38:37Z

Changes with this PR

New article in guide section "Microbots : Introduction, Installation Guide and Creating Your First MicroBot"
Created index file in guide section

Creation of new sections in Guide
Bots. Tools, Advanced
Fix of paths

From : microbots/copilot-bot/
To : microbots/guides/bots/copilot-bot/

From : microbots/blog/guides/tesseract_ocr_tool_use/
To : microbots/guides/tools/tesseract_ocr_tool_use/

From : /microbots/authentication/
To : microbots/guides/advanced/authentication/

codecov-commenter · 2026-04-17T11:40:12Z

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 97.08%. Comparing base (a080fd3) to head (41cb5a2).

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #132      +/-   ##
==========================================
- Coverage   97.15%   97.08%   -0.07%     
==========================================
  Files          30       30              
  Lines        1546     1546              
==========================================
- Hits         1502     1501       -1     
- Misses         44       45       +1

Flag	Coverage Δ
integration	`51.81% <ø> (+0.25%)`	⬆️
ollama_local	`48.18% <ø> (ø)`
slow-browser	`40.10% <ø> (ø)`
slow-other	`56.46% <ø> (-1.36%)`	⬇️
unit	`88.22% <ø> (ø)`

Flags with carried forward coverage won't be shown. Click here to find out more.
see 1 file with indirect coverage changes

🚀 New features to boost your workflow:

❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

0xba1a · 2026-04-20T10:26:01Z

@@ -0,0 +1,9 @@
+"use strict";


We don't need to add output files and log files. Only initial source files are sufficient as it can be used by the reader

Removed the build.log and app.js which was output of tsc app.ts

0xba1a · 2026-04-20T10:29:24Z

+!!! note
+    For advanced authentication options (Azure AD tokens, managed identity, service principals), see the [Authentication Guide](../advanced/authentication.md).
+
+## Install Microbots


This needs detailed walkthrough - Python installation, venv creation and microbots installation. Also, in Windows, we need to do it inside wsl.

Added python installation in Install Microbots prerequisites Section

0xba1a · 2026-04-20T10:31:43Z

+
+### Step 1: Prepare a sample project
+
+Create a project folder with a TypeScript file that has a syntax error. When the TypeScript compiler (`tsc`) tries to convert it to JavaScript, it fails — and the error output is captured in a log file.


User can create it or they can simply clone your example folder and follow the instructions step-by-step

Added detailed instruction on creating project and create files inside and expected project structure

0xba1a · 2026-04-20T10:32:03Z

+
+Create `code/app.ts` with a deliberate syntax error:
+
+```typescript title="code/app.ts"


Please use line number feature of snippet plugin

Added line number feature in code/app.ts

0xba1a · 2026-04-20T10:33:32Z

+
+```bash title="Terminal"
+cd code
+tsc app.ts > build.log 2>&1


Log files are naturally ignored by .gitignore. So, it will not be visible after deployment. Because sharing original log file may explose system related information. So, please use snippet of the log file here.

Removed build.log reference

0xba1a · 2026-04-20T10:33:53Z

+This produces the following build log:
+
+```log title="code/build.log"
+--8<-- "docs/examples/microbots_introduction/code/build.log"


build.log will not be checke-in to the git repo

Removed build.log reference

0xba1a · 2026-04-20T10:34:59Z

+    └── build.log
+```
+
+### Step 2: Analyze logs with a LogAnalysisBot


Considering this is going to be the very first blog for the user, we need to add more details here. The flow of agent creation and agent run. Individual arguments of Agent initialization and run function has to be explained.

Added detailed code walkthrough which can be expanded which will have each function call and arguments which it will accept ( mainly the Microbots Instance creation and run function )

0xba1a · 2026-04-20T10:36:57Z

+
+1. **Created a Docker container** with the `code` folder mounted using the appropriate permissions.
+2. **Sent your task** to the LLM along with a system prompt tailored to the bot type.
+3. **Executed commands** inside the container as directed by the LLM (e.g., `cat`, `grep`, `sed`).


Please import logging and set logging config to print INFO log in the LogAnalysisBot.py. So, user can see the individual command the LLM runs. Also, you can point them how to read the logs with some examples here.

Enabled logging info in code and added the full log output when user is interested to see

0xba1a · 2026-04-21T08:42:13Z

+
+**Install Python:**
+
+- **Windows:** Download the latest installer from [python.org](https://www.python.org/downloads/). During installation, check **"Add Python to PATH"**.


Did you verify running microbots on Windows without WSL. If not, you can only mention this guide for Linux setup. We'll update it soon for Windows setup.

0xba1a · 2026-04-21T08:43:30Z

+    sudo apt update && sudo apt install python3 python3-pip python3-venv
+
+    # Fedora
+    sudo dnf install python3 python3-pip


python3-venv is missing for Fedora. Please create a simple Fedora container and verify your steps whether they are valid.

0xba1a · 2026-04-21T08:45:09Z

+Create a `.env` file in the root of your application with:
+
+```env title=".env"
+OPEN_AI_END_POINT=https://your-resource-name.openai.azure.com


With reference Kavya's internal chat, END_POINT value should be enclosed in double quotes. Please verify.

0xba1a · 2026-04-21T08:48:06Z

+├── log_analysis_bot.py
+└── code/
+    ├── app.ts
+    ├── app.js


Will app.js present if ts file compilation fails?

0xba1a · 2026-04-21T08:50:06Z

+
+from dotenv import load_dotenv
+
+load_dotenv()


load_dotenv() will be used if environment variables are stored in a file named .env. As in your example, you are explicitely setting env variables, you may not need .env

0xba1a · 2026-04-21T08:54:03Z

+    |----------|------|----------|-------------|
+    | `model` | `str` | Yes | The LLM to use, in the format `<provider>/<model_name>`. Supported providers include `azure-openai`, `anthropic`, and `ollama`. |
+    | `folder_to_mount` | `str` | Yes | Path to the local folder the bot can access. For `LogAnalysisBot`, this is mounted as **read-only** inside the Docker container so the bot can read source files but never modify them. |
+    | `environment` | `any` | No | A custom execution environment. Defaults to a `LocalDockerEnvironment` if not provided. |


You don't need to mention all the possible arguments here. Please explain only those arguments. In parallel, you can generate an auto documentation and link to the class definition of LogAnalysisBot.

0xba1a · 2026-04-21T08:59:56Z

+    | Argument | Type | Default | Description |
+    |----------|------|---------|-------------|
+    | `file_name` | `str` | — | **(Required)** Path to the log file to analyze. The file is copied into the container as read-only. |
+    | `max_iterations` | `int` | `20` | Maximum number of LLM reasoning/command loops before the bot stops. Each iteration is one round of the LLM issuing a command and receiving its output. |


max_iteration can be removed from here as it is not used in the code

0xba1a · 2026-04-21T09:01:42Z

+
+    When `run()` is called, it:
+
+    1. Copies the log file into the Docker container at a dedicated `/var/log/` path.


This detailed explanation will not make sense until someone understood the agentic architecture and Microbot's internals. So, you can simply mention that the Agent is invoked by run. Once agent completes, the result will be available in result.result.

0xba1a · 2026-04-21T09:02:48Z

+    ```
+    Prints the bot's root-cause analysis.
+
+The `my_bot.run()` method returns a `BotRunResult` object with the following fields:


BotRunResult should be a link to the BotRunResult class documentation. In python document generation is easier. That document you should create and make corresponding link.

0xba1a · 2026-04-21T09:04:50Z

+```text title="Terminal output"
+(.venv) $ python3 log_analysis_bot.py
+
+INFO:microbots.environment.local_docker.LocalDockerEnvironment:🗂️  Created working directory at /home/sikannan/MICROBOTS_WORKDIR_18da47d7


Giving the complete terminal output is not advised as it may reveal some machine related secrets. You can simply mention how to run the code, where the user can find the log and how to interpret the log with snippets of log

microbots intro

9a1040a

shivashanmugam changed the title ~~microbots intro~~ New Guide Article : "Microbots : Introduction, Installation Guide and Creating Your First MicroBot" and structural changes in blog Apr 17, 2026

shivashanmugam requested a review from 0xba1a April 17, 2026 12:39

0xba1a requested changes Apr 20, 2026

View reviewed changes

shivashanmugam added 3 commits April 20, 2026 16:55

removing app.js file from the examples folder

31097a8

fix pr comments

8eaac5d

remove the os import

41cb5a2

shivashanmugam requested a review from 0xba1a April 21, 2026 08:50

0xba1a requested changes Apr 21, 2026

View reviewed changes


		### Step 1: Prepare a sample project

		Create a project folder with a TypeScript file that has a syntax error. When the TypeScript compiler (`tsc`) tries to convert it to JavaScript, it fails — and the error output is captured in a log file.


		Create `code/app.ts` with a deliberate syntax error:

		```typescript title="code/app.ts"


		Install Python:

		- Windows: Download the latest installer from [python.org](https://www.python.org/downloads/). During installation, check "Add Python to PATH".


		When `run()` is called, it:

		1. Copies the log file into the Docker container at a dedicated `/var/log/` path.

Conversation

shivashanmugam commented Apr 17, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Changes with this PR

Uh oh!

codecov-commenter commented Apr 17, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

shivashanmugam Apr 21, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

shivashanmugam Apr 21, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

shivashanmugam commented Apr 17, 2026 •

edited

Loading

codecov-commenter commented Apr 17, 2026 •

edited

Loading

shivashanmugam Apr 21, 2026 •

edited

Loading

shivashanmugam Apr 21, 2026 •

edited

Loading