Skip to content

Quick Start: Connect Toolkits

Learn how to connect external tools and services to ELITEA through toolkits. This guide covers the most commonly used integrations: Jira, Confluence, GitHub, Figma, TestRail, and Artifact storage.

What are Toolkits?

Toolkits connect ELITEA to external services, enabling you to:

  • Fetch data from project management tools
  • Search documentation
  • Analyze code repositories
  • Read design files
  • Manage test cases
  • Store and retrieve files

Before You Begin

Prerequisites

Requirements vary by toolkit type. Depending on the service you're connecting to, you may need:

  • Credentials: Account credentials for the external service
  • Authentication: API tokens, access keys, or passwords
  • Service URLs: Base URLs for self-hosted or enterprise services
  • Service-Specific Info: Repository names, space keys, project IDs, bucket names, etc.
  • Permissions: Appropriate access rights in the external service

Refer to each toolkit section below for specific requirements.

Jira Toolkit

Connect to Jira for issue tracking, project management, and agile workflows.

What You'll Need

  • Jira Base URL (e.g., https://yourcompany.atlassian.net)
  • Authentication (choose one):
    • Basic Auth: Username + API Key(server)
    • Bearer Token: Personal Access Token(cloud)

How to Connect

  1. Access Chat: Click Chat in the left sidebar

  2. Create a Toolkit from Chat:

    • In the PARTICIPANTS panel on the right, find toolkits
    • Click the + Add toolkit icon
    • Click + Create new Toolkit
    • Select Jira as the toolkit type
    • Enter a name (e.g., MyJiraToolkit)

  3. Create Credentials:

    • In the CONFIGURATION section, click the credentials dropdown
    • Select + New credentials (or + New private credentials)
    • You'll be redirected to the Create Credentials page
    • Select Jira as the credential type
    • Give it a name like My Jira Credentials
    • Fill in:
      • Base URL: Your Jira instance URL
      • For Basic Auth: Username + API Key
      • For Bearer Token: Token only
    • Click Save

  4. Complete Toolkit Configuration:

    • Return to the toolkit canvas (or refresh if needed)
    • Click the refresh icon next to the Jira configuration field
    • Select your newly created credential from the dropdown
    • Select an Embedding Model (e.g., text-embedding-3-small)
    • Choose a Vector Store (PostgreSQL with pgvector extension)
    • Enable the tools you need (e.g., Search using JQL, Create issue)
    • Click Create
    • Close the Canvas interface

  5. Use it:

    • Type: Find all open bugs in project ABC
    • The toolkit will search Jira and return results

Connect Toolkit

Common Tools

  • Search using JQL: Query issues with Jira Query Language
  • Create issue: Create new Jira issues
  • Update issue: Modify existing issues
  • Set issue status: Change issue workflow status
  • Add comments: Add comments to issues
  • Get comments with image descriptions: Retrieve comments including AI-generated image descriptions
  • Link issues: Create relationships between issues
  • Modify labels: Add or remove issue labels
  • Get attachments content: Download and access file attachments
  • List projects: View available Jira projects
  • Search index: Perform semantic search using vector embeddings (requires indexing)

Confluence Toolkit

Connect to Confluence for documentation search and knowledge base access.

What You'll Need

  • Confluence Base URL (e.g., https://yourcompany.atlassian.net/wiki)
  • Authentication (choose one):
    • Basic Auth: Username + API Key(server)
    • Bearer Token: Personal Access Token(cloud)

How to Connect

  1. Access Chat: Click Chat in the left sidebar

  2. Create a Toolkit from Chat:

    • In PARTICIPANTS panel, click + Add toolkit
    • Click + Create new Toolkit
    • Select Confluence

  3. Create Credentials:

    • Click credentials dropdown → + New credentials
    • Select Confluence as credential type
    • Name it (e.g., My Confluence Credentials)
    • Fill in:
      • Base URL: Your Confluence instance URL
      • For Basic Auth: Username + API Key
      • For Bearer Token: Token only
    • Click Save

  4. Complete Toolkit Configuration:

    • Return to toolkit canvas
    • Refresh and select your credential
    • Specify a Confluence space to limit searches-the toolkit will be automatically named with this space name
    • Select an Embedding Model (e.g., text-embedding-3-small)
    • Choose a Vector Store (PostgreSQL with pgvector extension)
    • Enable the tools you need (e.g., Read page by id, Update page by title)
    • Click Create

  5. Use it:

    • Type: Search Confluence for API documentation
    • The toolkit will search and return relevant pages

Confluence

Common Tools

  • Search pages: Find pages across Confluence
  • Site search: Perform full-text search across all content
  • Read page by id: Retrieve specific page content
  • Get page with image descriptions: Get page content including AI-generated image descriptions
  • Create pages: Create new Confluence pages
  • Update page by id: Modify existing pages by ID
  • Update page by title: Modify existing pages by title
  • Get page attachments: Access page attachments
  • Get pages with label: Retrieve pages by label
  • Search by title: Find pages by title
  • Search index: Perform semantic search using vector embeddings (requires indexing)

GitHub Toolkit

Connect to GitHub repositories for code analysis, issue management, and pull request operations.

What You'll Need

  • Base URL (default: https://api.github.com, or your GitHub Enterprise URL)
  • Authentication (choose one):
    • Access Token: Personal Access Token (recommended)
    • Username + Password: Basic authentication
    • App Private Key: GitHub App ID + Private Key
    • Anonymous: No authentication (public repositories only)

How to Connect

  1. Access Chat: Click Chat in the left sidebar

  2. Create a Toolkit from Chat:

    • In PARTICIPANTS panel, click + Add toolkit
    • Click + Create new Toolkit
    • Select GitHub

  3. Create Credentials (optional for public repos):

    • Click credentials dropdown → + New credentials
    • Select GitHub as credential type
    • Name it (e.g., My GitHub Credentials)
    • Fill in (choose one method):
      • Token: Access Token only
      • Password: Username + Password
      • App Private Key: App ID + Private Key
    • Click Save

  4. Complete Toolkit Configuration:

    • Return to toolkit canvas
    • Refresh and select your credential (or leave empty for public access)
    • Configure:
      • Project: Organization or username
      • Repository: Repository name(e.g., ProjectAlita/projectalita.github.io)
      • Branch: Default branch (e.g., main)
    • Select an Embedding Model (e.g., text-embedding-3-small)
    • Choose a Vector Store (PostgreSQL with pgvector extension)
    • Enable the tools you need (e.g., Read file, List pull request diffs)
    • Click Create

  5. Use it:

    • Type: Get the guide from projectalita.github.io repo for creating secrets in ELITEA
    • The toolkit will fetch and display the secret creation documentation

    GitHub

Common Tools

  • Read file: Read files from repository
  • List files in bot branch: Browse repository contents
  • Create issue: Create new GitHub issues
  • Update issue: Modify existing issues
  • Comment on issue: Add comments to issues
  • Get issue: Retrieve issue details
  • List open pull requests: View open PRs
  • List pull request diffs: View changes in PRs
  • Get commits: View commit history
  • Create branch: Create new branches
  • Update file: Modify repository files
  • Search index: Perform semantic search using vector embeddings (requires indexing)

Figma Toolkit

Connect to Figma for design file access and collaboration.

What You'll Need

  • Figma Token: Personal Access Token from Figma account settings

How to Connect

  1. Access Chat: Click Chat in the left sidebar

  2. Create a Toolkit from Chat:

    • In PARTICIPANTS panel, click + Add toolkit
    • Click + Create new Toolkit
    • Select Figma
    • Enter a name (e.g., MyFigmaToolkit)

  3. Create Credentials:

    • Click credentials dropdown → + New credentials
    • Select Figma as credential type
    • Name it (e.g., My Figma Credentials)
    • Fill in:
      • Token: Your Figma Personal Access Token
    • Click Save

  4. Complete Toolkit Configuration:

    • Return to toolkit canvas
    • Refresh and select your credential
    • Select an Embedding Model (e.g., text-embedding-3-small)
    • Choose a Vector Store (PostgreSQL with pgvector extension)
    • Enable tools for accessing Figma files(e.g. Get file, Get project files)
    • Click Create

  5. Use it:

    • Type: Get information from Figma file XYZ
    • The toolkit will access Figma and retrieve file data

    Figma

Getting Your Figma Token

  1. Go to Figma account settings
  2. Navigate to Personal Access Tokens
  3. Click Generate new token
  4. Copy the token immediately (it won't be shown again)

Common Tools

  • Get file: Retrieve Figma file information
  • Get file images: Extract images from Figma files
  • Get file nodes: Access specific nodes within files
  • Get file comments: View comments on designs
  • Post file comment: Add comments to Figma files
  • Get file versions: View version history
  • Get project files: List files in a project
  • Get team projects: View team projects
  • Search index: Perform semantic search using vector embeddings (requires indexing)

TestRail Toolkit

Connect to TestRail for test case management and test execution tracking.

What You'll Need

  • TestRail URL (e.g., https://yourcompany.testrail.io)
  • Email: Your TestRail account email
  • Password: Your TestRail password or API key

How to Connect

  1. Access Chat: Click Chat in the left sidebar

  2. Create a Toolkit from Chat:

    • In PARTICIPANTS panel, click + Add toolkit
    • Click + Create new Toolkit
    • Select TestRail

  3. Create Credentials:

    • Click credentials dropdown → + New credentials
    • Select TestRail as credential type
    • Name it (e.g., My TestRail Credentials)
    • Fill in:
      • URL: Your TestRail instance URL (e.g., https://yourcompany.testrail.io)
      • Email: Your account email (e.g., user@company.com)
      • Password: Your password or API key
    • Click Save

  4. Complete Toolkit Configuration:

    • Return to toolkit canvas
    • Refresh and select your credential
    • Enter a name (e.g., MyTestRailToolkit)
    • Select an Embedding Model (e.g., text-embedding-3-small)
    • Choose a Vector Store (PostgreSQL with pgvector extension)
    • Enable the tools you need (e.g., Get cases, Add case)
    • Click Create

  5. Use it:

    • Type: Show me all test cases related to log-in functionality from project with id 1
    • The toolkit will fetch and display test case information

    Testrail

Common Tools

  • Get cases: Retrieve test case information
  • Get case: Get specific test case details
  • Get cases by filter: Filter and search test cases
  • Add case: Create new test case
  • Add cases: Create multiple test cases
  • Update case: Modify existing test cases
  • Get suites: List test suites
  • Search index: Perform semantic search using vector embeddings (requires indexing)

Artifact Toolkit

Create storage buckets for file attachments in chat conversations.

What You'll Need

  • Bucket Name: A unique identifier (lowercase letters, numbers, hyphens only)

When to Use Artifact Toolkit

The Artifact Toolkit is essential for:

  • Enabling file attachments in chat
  • Storing uploaded images and documents
  • Managing conversation artifacts
  • File retention and organization

How to Connect

Manual Setup from Toolkit Canvas:

  1. Access Chat: Click Chat in the left sidebar

  2. Create a Toolkit:

    • In PARTICIPANTS panel, click + Add toolkit
    • Click + Create new Toolkit
    • Select Artifact
    • Enter a name that matches your bucket name

  3. Configure Bucket:

    • Bucket: Enter a unique name (lowercase, alphanumeric, hyphens)
      • ✔️ Valid: my-files, project-2024, team-assets
      • ✘ Invalid: My Files, project_2024, Team@Assets
    • Select an Embedding Model (e.g., text-embedding-3-small)
    • Choose a Vector Store (PostgreSQL with pgvector extension)
    • Enable the tools you need (e.g., CreateFile, ListFiles)
    • Click Create

  4. Use it:

    • Files uploaded via chat are stored in this bucket
    • Agents can read, create, and manage files programmatically
    • Example: List all files in my artifact bucket or Read the contents of report.pdf

    Artifact

Common Tools

  • CreateFile: Write new files to bucket
  • ReadFile: Read file contents
  • ListFiles: Browse bucket contents
  • DeleteFile: Remove files
  • AppendData: Add data to existing files
  • OverwriteData: Replace file contents
  • CreateNewBucket: Create additional storage buckets
  • Search index: Perform semantic search using vector embeddings (requires indexing)

Quick Setup from Chat

The easiest way to create an Artifact Toolkit is directly from chat when attaching files. See Send an Attachment.

Troubleshooting

Authentication Errors (401/403)

  • Verify credentials are correct
  • Check API token hasn't expired
  • Ensure user has required permissions
  • For Atlassian tools, verify correct authentication method

Connection Issues

  • Verify base URL is correct** (must include http:// or https://)
  • Check firewall/proxy settings
  • Test connection from browser first
  • Ensure service is accessible from your network

Toolkit Not Appearing

  • Click refresh icon next to credential selector
  • Verify toolkit was created successfully
  • Check project permissions
  • Reload the page

Tools Return No Results

  • Check the "Limit" field in toolkit configuration
  • Verify search parameters are correct
  • Ensure you have access to the requested data
  • Check service-specific query syntax (e.g., JQL for Jira)

Pro Tips

  • Use private credentials for personal API tokens to keep them secure
  • Use secrets to securely store sensitive data like API keys and
  • Create separate toolkits for different projects or contexts
  • Test connections using the "Test Connection" button in credentials
  • Start with read-only tools before enabling create/update operations
  • Use descriptive names for toolkits and credentials to stay organized

Useful Links