gaunt-sloth-assistant 0.1.1 → 0.1.3

package/.github/workflows/ci.yml

@@ -0,0 +1,33 @@
+name: Tests and Lint
+
+on:
+  push:
+    branches: [ "main" ]
+  pull_request:
+    branches: [ "main" ]
+
+jobs:
+  test-and-lint:
+    runs-on: ubuntu-latest
+    
+    strategy:
+      matrix:
+        node-version: [22.x]
+
+    steps:
+    - uses: actions/checkout@v4
+    
+    - name: Use Node.js ${{ matrix.node-version }}
+      uses: actions/setup-node@v4
+      with:
+        node-version: ${{ matrix.node-version }}
+        cache: 'npm'
+    
+    - name: Install dependencies
+      run: npm ci
+    
+    - name: Run ESLint
+      run: npm run lint
+    
+    - name: Run Tests
+      run: npm test

package/.gsloth.preamble.review.md

@@ -18,6 +18,7 @@ You are conducting LangChain.js/LangGraph.js Node.JS code review.
 
 ### Security
 - Validate API key handling and environment variables
+- Make sure no personal data is present in code
 - Check for proper input sanitization
 - Review rate limiting implementation
 - Verify output validation and sanitization

package/README.md

@@ -1,7 +1,33 @@
 # Gaunt Sloth Assistant
-Simplistic AI assistant helping to do **code reviews from command line** based on [Langchain.js](https://github.com/langchain-ai/langchainjs)
-
-## Review PR (Pull Request)
+[![Tests and Lint](https://github.com/andruhon/gaunt-sloth-assistant/actions/workflows/ci.yml/badge.svg?event=push)](https://github.com/andruhon/gaunt-sloth-assistant/actions/workflows/ci.yml)
+
+Gaunt Sloth Assistant is a Simplistic **command line AI assistant**
+for software developers,
+who wish to reduce cognitive load and time spending on **code reviews** (and pull request diff reviews).
+
+Based on [Langchain.js](https://github.com/langchain-ai/langchainjs)
+
+## What GSloth does:
+- Reviews code;
+  - Suggests bug fixes;
+  - Explains provided code
+- Reviews Diffs provided with pipe (|);
+  - You can ask GSloth to review your own code before committing.
+- Reviews Pull Requests (PRs);
+  - Fetches descriptions (requirements) from Jira;
+- Answers questions about provided code;
+- Writes code;
+- Saves all responses to the project directory;
+- Anything else you need, when combined with other command line tools.
+
+### To make GSloth work, you need an **API key** from some AI provider, such as:
+- Google Vertex AI;
+- Anthropic;
+- Groq.
+
+## Primary Functions:
+
+### Review PR (Pull Request)
 To review PR by PR number:
 
 First make sure the official [GitHub cli (gh)](https://cli.github.com/) is installed
@@ -19,13 +45,8 @@ Review providing markdown file with requirements and notes.
 ```shell
 gsloth pr 42 -f PROJ-1234.md
 ```
-Jira integration is in [ROADMAP](ROADMAP.md).
-Currently, the easiest ***meaningful*** way to add jira description is to
-open Jira XML with "Export XML" in jira and to copy `<description></description>` block.
-This block contains HTML and AI understands it easily 
-(most importantly it understand nested lists like ul>li).
 
-## JIRA Integration
+### JIRA Integration
 
 When JIRA integration is configured, the JIRA issue text can be included alongside the diff for review.
 The project review preamble can be modified to reject a pull request immediately
@@ -39,31 +60,26 @@ supplies description of JIRA issue with number PP-4242:
 gsloth pr 42 PP-4242
 ```
 
-Example configuration setting up JIRA integration using a legacy API token.
+Example configuration setting up JIRA integration using a legacy API token for both `review` and `pr` commands.
 Make sure you use your actual company domain in `baseUrl` and your personal legacy `token`.
 
 A legacy token can be acquired from `Atlassian Account Settings -> Security -> Create and manage API tokens`.
 
-```javascript
-export async function configure(importFunction, global) {
-    const vertexAi = await importFunction('@langchain/google-vertexai');
-    return {
-        llm: new vertexAi.ChatVertexAI({
-            model: "gemini-2.5-pro-exp-03-25"
-        }),
-        requirementsProvider: 'jira-legacy',
-        requirementsProviderConfig: {
-            'jira-legacy': {
-                username: 'andrei.kondratev@unimarket.com', // Your Jira username/email
-                token: 'YOURSECRETTOKEN',     // Replace with your real Jira API token
-                baseUrl: 'https://yourcompany.atlassian.net/rest/api/2/issue/'  // Your Jira instance base URL
-            }
-        }
+```json
+{
+  "llm": {"type": "vertexai", "model": "gemini-2.5-pro-exp-03-25"},
+  "requirementsProvider": "jira-legacy",
+  "requirementsProviderConfig": {
+    "jira-legacy": {
+      "username": "user@yourcompany.com",
+      "token": "YOUR_JIRA_LEGACY_TOKEN",
+      "baseUrl": "https://yourcompany.atlassian.net/rest/api/2/issue/"
     }
+  }
 }
 ```
 
-## Review any Diff
+### Review any Diff
 ```shell
 git --no-pager diff origin/master...yourgitcommithash | gsloth review
 ```
@@ -74,7 +90,7 @@ Review current local changes:
 git --no-pager diff | gsloth review
 ```
 
-## Question Answering
+### Question Answering
 ```shell
 gsloth ask "which types of primitives are available in JavaScript?"
 ```
@@ -83,17 +99,30 @@ gsloth ask "which types of primitives are available in JavaScript?"
 gsloth ask "Please have a look at this file" -f index.js
 ```
 
+Multiple files may be provided as well
+
+```shell
+gsloth ask "Please have a look at these files" -f index.js test.js
+```
+
 ## Installation
 
 Tested with Node 22 LTS.
 
-## NPM
+### NPM
 ```shell
 npm install gaunt-sloth-assistant -g
 ```
 
 ## Configuration
-Go to your project directory and init sloth with vendor of your choice.
+
+> Gaunt Sloth currently only functions from the directory which has a configuration file (`.gsloth.config.js`, `.gsloth.config.json`, or `.gsloth.config.mjs`) and `.gsloth.preamble.review.md`.
+> Global configuration to invoke gsloth anywhere is in [ROADMAP](ROADMAP.md).
+
+Configuration can be created with `gsloth init [vendor]` command.
+Currently, vertexai, anthropic and groq can be configured with `gsloth init [vendor]`.
+
+More detailed information on configuration can be found in [CONFIGURATION.md](./docs/CONFIGURATION.md)
 
 ### Google Vertex AI
 ```shell
@@ -104,92 +133,26 @@ gcloud auth application-default login
 ```
 
 ### Anthropic
+
 ```shell
 cd ./your-project
 gsloth init anthropic
 ```
-Make sure you either define `ANTHROPIC_API_KEY` environment variable or edit `.gsloth.config.js` and set up your key.
+
+Make sure you either define `ANTHROPIC_API_KEY` environment variable or edit your configuration file and set up your key.
 
 ### Groq
 ```shell
 cd ./your-project
 gsloth init groq
 ```
-Make sure you either define `GROQ_API_KEY` environment variable or edit `.gsloth.config.js` and set up your key.
-
-### Further configuration
-
-Currently vertexai, anthropic and groq can be configured with `gsloth init`.
-
-Populate `.gsloth.preamble.review.md` with your project details and quality requirements.
-Proper preamble is a paramount for good inference.
-Check [.gsloth.preamble.review.md](.gsloth.preamble.review.md) for example.
-
-### Manual configuration.
-Your project should have the following files in order for gsloth to function:
-- `.gsloth.config.js`
-- `.gsloth.preamble.review.md`
-
-Global configuration to invoke gsloth anywhere is in [ROADMAP](ROADMAP.md).
-
-**Example of .gsloth.config.js for Anthropic**  
-```javascript
-export async function configure(importFunction, global) {
-    // this is going to be imported from sloth dependencies,
-    // but can potentially be pulled from global node modules or from this project
-    // At a moment only google-vertexai and anthropic packaged with Sloth, but you can install support for any other langchain llms
-    const anthropic = await importFunction('@langchain/anthropic');
-    return {
-        llm: new anthropic.ChatAnthropic({
-            apiKey: process.env.ANTHROPIC_API_KEY, // Default value, but you can provide the key in many different ways, even as literal
-            model: "claude-3-5-sonnet-20241022"
-        })
-    };
-}
-```
+Make sure you either define `GROQ_API_KEY` environment variable or edit your configuration file and set up your key.
 
-**Example of .gsloth.config.js for VertexAI**  
-VertexAI usually needs `gcloud auth application-default login`
-(or both `gcloud auth login` and `gcloud auth application-default login`) and does not need any separate API keys.
-```javascript
-export async function configure(importFunction, global) {
-    // this is going to be imported from sloth dependencies,
-    // but can potentially be pulled from global node modules or from this project
-    // At a moment only google-vertexai and anthropic packaged with Sloth, but you can install support for any other langchain llms
-    // Note: for vertex AI you likely to need to do `gcloud auth login`
-    const vertexAi = await importFunction('@langchain/google-vertexai');
-    return {
-        llm: new vertexAi.ChatVertexAI({
-            model: "gemini-2.5-pro-exp-03-25", // Consider checking for latest recommended model versions
-            temperature: 0,
-            //// Other parameters might be relevant depending on Vertex AI API updates.
-            //// The project is not in the interface, but it is in documentation and it seems to work.
-            // project: 'your-cool-google-cloud-project',
-        })
-    }
-}
-```
+### Other AI providers
+Any other AI provider supported by Langchain.js can be configured with js [Config](./docs/CONFIGURATION.md). 
 
-**Example of .gsloth.config.js for Groq**  
-VertexAI usually needs `gcloud auth application-default login`
-(or both `gcloud auth login` and `gcloud auth application-default login`) and does not need any separate API keys.
-```javascript
-export async function configure(importFunction, global) {
-    // this is going to be imported from sloth dependencies,
-    // but can potentially be pulled from global node modules or from this project
-    const groq = await importFunction('@langchain/groq');
-    return {
-        llm: new groq.ChatGroq({
-            model: "deepseek-r1-distill-llama-70b", // Check other models available
-            apiKey: process.env.GROQ_API_KEY, // Default value, but you can provide the key in many different ways, even as literal
-        })
-    };
-}
-```
-
-The configure function should simply return instance of langchain [chat model](https://v03.api.js.langchain.com/classes/_langchain_core.language_models_chat_models.BaseChatModel.html).
-See [Langchain documentation](https://js.langchain.com/docs/tutorials/llm_chain/) for more details.
+## Building from repo
+See [DEVELOPMENT.md](./docs/DEVELOPMENT.md)
 
 ## License
 License is [MIT](https://opensource.org/license/mit). See [LICENSE](LICENSE)
-

package/ROADMAP.md

@@ -1,28 +1,28 @@
 # Gaunt Sloth Assistant roadmap
 
-
 ## 1.0.0
 Doing the following below and making it work stably should be sufficient to call it version 1. 
 
-### ⌛Add tests and gain reasonable coverage
-### Configure eslint for code quality checks
-### Automate release process
-### Add project init command
-Add a command to init certain model in certain project, for example `gsloth init gemini`
-or `gsloth init` and select one of the provided options. 
--[x] VertexAI
--[x] Anthropic
--[x] Groq
--[ ] Local LLm 
-### Allow global configuration
-### Streamline and stabilize configuration
-### Add JIRA legacy token integration plugin
-### Teach assistant to identify important files and include their contents into prompt
-The idea is to ask smaller model like flash to find important files from diff then pick them up and include into prompt.
-### Teach assistant to access provided public web links
-### Consider adding an option to always include certain source code files into prompt
-### Test with Groq
-### Add general chat command
+### Checklist
+
+- [x] Add tests and gain reasonable coverage
+- [x] Configure eslint for code quality checks
+- [ ] Automate release process
+- [x] Add project init command
+  Add a command to init certain model in certain project, for example `gsloth init gemini`
+  or `gsloth init` and select one of the provided options. 
+  -[x] VertexAI
+  -[x] Anthropic
+  -[x] Groq
+  -[ ] Local LLm 
+- [ ] Allow global configuration
+- [ ] Streamline and stabilize configuration
+- [ ] Add JIRA legacy token integration plugin
+- [ ] Teach assistant to identify important files and include their contents into prompt
+  The idea is to ask smaller model like flash to find important files from diff then pick them up and include into prompt.
+- [ ] Teach assistant to access provided public web links
+- [ ] Consider adding an option to always include certain source code files into prompt
+- [ ] Add general chat command
 
 ## Extra stuff for later
 

package/UX-RESEARCH.md

@@ -34,7 +34,7 @@ Arguments:
   prNumber           PR number to review
 
 Options:
-  -f, --file <file>  Input file. Context of this file will be added BEFORE the
+  -f, --file <file>  Input file. Content of this file will be added BEFORE the
                      diff
   -h, --help         display help for command
 ```

package/docs/CONFIGURATION.md

@@ -0,0 +1,190 @@
+# Configuration
+
+Populate `.gsloth.preamble.review.md` with your project details and quality requirements.
+Proper preamble is a paramount for good inference.
+Check [.gsloth.preamble.review.md](../.gsloth.preamble.review.md) for example.
+
+Your project should have the following files in order for gsloth to function:
+- Configuration file (one of):
+  - `.gsloth.config.js` (JavaScript module)
+  - `.gsloth.config.json` (JSON file)
+  - `.gsloth.config.mjs` (JavaScript module with explicit module extension)
+- `.gsloth.preamble.review.md`
+
+> Gaunt Sloth currently only functions from the directory which has one of the configuration files and `.gsloth.preamble.review.md`.
+> Global configuration to invoke gsloth anywhere is in [ROADMAP](../ROADMAP.md).
+
+## Config initialization
+Configuration can be created with `gsloth init [vendor]` command.
+Currently, vertexai, anthropic and groq can be configured with `gsloth init [vendor]`.
+
+### Google Vertex AI
+```shell
+cd ./your-project
+gsloth init vertexai
+gcloud auth login
+gcloud auth application-default login
+```
+
+### Anthropic
+```shell
+cd ./your-project
+gsloth init anthropic
+```
+Make sure you either define `ANTHROPIC_API_KEY` environment variable or edit your configuration file and set up your key.
+
+### Groq
+```shell
+cd ./your-project
+gsloth init groq
+```
+Make sure you either define `GROQ_API_KEY` environment variable or edit your configuration file and set up your key.
+
+## Examples of configuration for different providers 
+
+### JSON Configuration (.gsloth.config.json)
+
+JSON configuration is simpler but less flexible than JavaScript configuration. It should directly contain the configuration object.
+
+**Example of .gsloth.config.json for Anthropic**
+```json
+{
+  "llm": {
+    "type": "anthropic",
+    "apiKey": "your-api-key-here",
+    "model": "claude-3-5-sonnet-20241022"
+  }
+}
+```
+
+**Example of .gsloth.config.json for VertexAI**
+```json
+{
+  "llm": {
+    "type": "vertexai",
+    "model": "gemini-2.5-pro-exp-03-25",
+    "temperature": 0
+  }
+}
+```
+
+**Example of .gsloth.config.json for Groq**
+```json
+{
+  "llm": {
+    "type": "groq",
+    "model": "deepseek-r1-distill-llama-70b",
+    "apiKey": "your-api-key-here"
+  }
+}
+```
+
+### JavaScript Configuration (.gsloth.config.js or .gsloth.config.mjs)
+
+**Example of .gsloth.config.js for Anthropic**
+```javascript
+export async function configure(importFunction, global) {
+    // this is going to be imported from sloth dependencies,
+    // but can potentially be pulled from global node modules or from this project
+    // At a moment only google-vertexai and anthropic packaged with Sloth, but you can install support for any other langchain llms
+    const anthropic = await importFunction('@langchain/anthropic');
+    return {
+        llm: new anthropic.ChatAnthropic({
+            apiKey: process.env.ANTHROPIC_API_KEY, // Default value, but you can provide the key in many different ways, even as literal
+            model: "claude-3-5-sonnet-20241022"
+        })
+    };
+}
+```
+
+**Example of .gsloth.config.js for VertexAI**  
+VertexAI usually needs `gcloud auth application-default login`
+(or both `gcloud auth login` and `gcloud auth application-default login`) and does not need any separate API keys.
+```javascript
+export async function configure(importFunction, global) {
+    // this is going to be imported from sloth dependencies,
+    // but can potentially be pulled from global node modules or from this project
+    // At a moment only google-vertexai and anthropic packaged with Sloth, but you can install support for any other langchain llms
+    // Note: for vertex AI you likely to need to do `gcloud auth login`
+    const vertexAi = await importFunction('@langchain/google-vertexai');
+    return {
+        llm: new vertexAi.ChatVertexAI({
+            model: "gemini-2.5-pro-exp-03-25", // Consider checking for latest recommended model versions
+            // API Key from AI Studio should also work
+            temperature: 0,
+            //// Other parameters might be relevant depending on Vertex AI API updates.
+            //// The project is not in the interface, but it is in documentation and it seems to work.
+            // project: 'your-cool-google-cloud-project',
+        })
+    }
+}
+```
+
+**Example of .gsloth.config.js for Groq**  
+VertexAI usually needs `gcloud auth application-default login`
+(or both `gcloud auth login` and `gcloud auth application-default login`) and does not need any separate API keys.
+```javascript
+export async function configure(importFunction, global) {
+    // this is going to be imported from sloth dependencies,
+    // but can potentially be pulled from global node modules or from this project
+    const groq = await importFunction('@langchain/groq');
+    return {
+        llm: new groq.ChatGroq({
+            model: "deepseek-r1-distill-llama-70b", // Check other models available
+            apiKey: process.env.GROQ_API_KEY, // Default value, but you can provide the key in many different ways, even as literal
+        })
+    };
+}
+```
+
+## Using other AI providers
+
+The configure function should simply return instance of langchain [chat model](https://v03.api.js.langchain.com/classes/_langchain_core.language_models_chat_models.BaseChatModel.html).
+See [Langchain documentation](https://js.langchain.com/docs/tutorials/llm_chain/) for more details.
+
+## Content providers
+
+### JIRA
+
+Example configuration setting up JIRA integration using a legacy API token for both `review` and `pr` commands.
+Make sure you use your actual company domain in `baseUrl` and your personal legacy `token`.
+
+A legacy token can be acquired from `Atlassian Account Settings -> Security -> Create and manage API tokens`.
+
+JSON:
+
+```json
+{
+  "llm": {"type": "vertexai", "model": "gemini-2.5-pro-exp-03-25"},
+  "requirementsProvider": "jira-legacy",
+  "requirementsProviderConfig": {
+    "jira-legacy": {
+      "username": "username@yourcompany.com",
+      "token": "YOUR_JIRA_LEGACY_TOKEN",
+      "baseUrl": "https://yourcompany.atlassian.net/rest/api/2/issue/"
+    }
+  }
+}
+```
+
+JavaScript:
+
+```javascript
+export async function configure(importFunction, global) {
+    const vertexAi = await importFunction('@langchain/google-vertexai');
+    return {
+        llm: new vertexAi.ChatVertexAI({
+            model: "gemini-2.5-pro-exp-03-25"
+        }),
+        requirementsProvider: 'jira-legacy',
+        requirementsProviderConfig: {
+            'jira-legacy': {
+                username: 'username@yourcompany.com', // Your Jira username/email
+                token: 'YOUR_JIRA_LEGACY_TOKEN',     // Replace with your real Jira API token
+                baseUrl: 'https://yourcompany.atlassian.net/rest/api/2/issue/'  // Your Jira instance base URL
+            }
+        }
+    }
+}
+```
+

package/docs/DEVELOPMENT.md

@@ -0,0 +1,21 @@
+# Installation
+
+## GitHub (master)
+
+Install dev version globally
+
+```shell
+git clone https://github.com/andruhon/gaunt-sloth.git
+npm install
+npm install -g ./
+```
+## Testing
+
+Unit tests are implemented with [Jasmine](https://github.com/jasmine/jasmine)
+and [testdouble.js](https://github.com/testdouble/testdouble.js).
+
+Running unit tests:
+
+```shell
+npm run test
+```

package/eslint.config.js

@@ -0,0 +1,38 @@
+import js from "@eslint/js";
+import { defineConfig } from "eslint/config";
+import jasmine from "eslint-plugin-jasmine"
+import globals from "globals";
+
+const { setTimeout, setInterval, clearInterval, Buffer, fetch } = globals.node;
+
+export default defineConfig([
+  {
+    files: ["src/**/*.js"],
+    plugins: {
+      js,
+    },
+    extends: ["js/recommended"],
+    rules: {
+      semi: "error",
+      "eol-last": "error"
+    },
+    languageOptions: {
+      globals: {
+        setTimeout,
+        setInterval,
+        clearInterval,
+        Buffer,
+        fetch
+      },
+    },
+  },
+  {
+    files: ["spec/**/*.js"],
+    plugins: { jasmine },
+    extends: ["jasmine/recommended"],
+    rules: {
+      semi: "error",
+      "eol-last": "error",
+    }
+  },
+]);

package/index.js

@@ -7,11 +7,13 @@ import { initCommand } from "./src/commands/initCommand.js";
 import { askCommand } from "./src/commands/askCommand.js";
 import { slothContext } from "./src/config.js";
 import { getSlothVersion, readStdin } from "./src/utils.js";
+import {getCurrentDir, getInstallDir, setInstallDir} from "./src/systemUtils.js";
 
 const program = new Command();
 
-slothContext.currentDir = process.cwd();
-slothContext.installDir = dirname(fileURLToPath(import.meta.url))
+setInstallDir(dirname(fileURLToPath(import.meta.url)));
+slothContext.currentDir = getCurrentDir();
+slothContext.installDir = getInstallDir();
 
 program
     .name('gsloth')

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "gaunt-sloth-assistant",
-  "version": "0.1.1",
+  "version": "0.1.3",
   "description": "",
   "license": "MIT",
   "author": "Andrew Kondratev",
@@ -12,14 +12,14 @@
     "npm": ">=10.9.0"
   },
   "scripts": {
-    "test": "jasmine"
+    "test": "jasmine",
+    "lint": "eslint"
   },
   "bin": {
     "gsloth": "index.js",
     "gth": "index.js"
   },
   "dependencies": {
-    "@eslint/js": "^9.25.0",
     "@langchain/anthropic": "^0.3.18",
     "@langchain/core": "^0.3.45",
     "@langchain/google-vertexai": "^0.2.4",
@@ -31,6 +31,10 @@
     "uuid": "^11.1.0"
   },
   "devDependencies": {
+    "@eslint/js": "^9.25.1",
+    "eslint": "^9.25.1",
+    "eslint-plugin-jasmine": "^4.2.2",
+    "globals": "^16.0.0",
     "jasmine": "^5.6.0",
     "testdouble": "^3.20.2"
   }

package/spec/.gsloth.config.js

@@ -18,5 +18,5 @@ export async function configure(importFunction, global) {
                 test: 'example'
             }
         }
-    }
+    };
 }