Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.jataka.io/llms.txt

Use this file to discover all available pages before exploring further.

​
Quick Start Guide

Use this guide to set up Jataka with GitHub, Salesforce, CI secrets, and Jira in a production-ready way.
1
Step 1 - Connect GitHubConnect your GitHub account from the Jataka dashboard. Authorize repository access for the organization you want to onboard.
2
Step 2 - Select GitHub RepositorySelect the GitHub repository where your Salesforce metadata and workflows are maintained. Confirm that the selected repository is the same one used for your release process.
3
Step 3 - Select Branch and BrainSelect the target branch that should run Jataka checks. If your workspace uses Brain profiles, choose the appropriate Brain for this repository.
4
Step 4 - Connect Salesforce OrgConnect Salesforce and choose the correct environment: production (login.salesforce.com) or sandbox (test.salesforce.com). Verify that the connected org matches the environment used by your deployment pipeline.
5
Step 5 - Approve Salesforce PermissionsAllow all the permissions
6
Step 6 - Generate SFDX Auth URLAuthenticate your target Salesforce org in the CLI first:
sf org login web --alias staging-org
Then display the org details and copy the generated SFDX auth URL:
sf org display --target-org staging-org --verbose
7
Step 7 - Add SFDX URL as GitHub SecretOptional: Add the generated auth URL as a GitHub Actions secret as SFDX_AUTH_URL. Add this to your workflow file to make sure changes are deployed the testing environment before running tests:
- name: Authenticate to Salesforce
  run: |
    echo "${{ secrets.SFDX_AUTH_URL }}" > auth_url.txt
    sf org login sfdx-url --sfdx-url-file auth_url.txt --set-default --alias staging-org
    rm auth_url.txt
8
Step 8 - Generate Jataka API KeyGenerate the Jataka API key from the integration setup panel (Step 3 flow). Save it in GitHub Secrets as JATAKA_API_KEY the same way as other credentials.
9
Step 9 - Add Workflow Command and Test ModeAdd this step to your project workflow file (for example, .github/workflows/deploy.yml) after Salesforce deployment:
- name: Trigger Jataka AI UI Tests
  # Put this step AFTER your Salesforce deployment step (Gearset, Copado, or SFDX)
  run: |
    curl -X POST "${{ secrets.JATAKA_API_URL }}/api/integrations/github/trigger" \
      -H "Authorization: Bearer ${{ secrets.JATAKA_API_KEY }}" \
      -H "Content-Type: application/json" \
      -d '{
        "installation_id": "YOUR_INSTALLATION_ID", 
        "repo_full_name": "${{ github.repository }}",
        "branch": "${{ github.head_ref || github.ref_name }}",
        "pr_number": ${{ github.event.pull_request.number || 'null' }},
        "test_mode": "${{ vars.JATAKA_TEST_MODE || 'auto' }}",
        "action": "${{ github.event.action }}",
        "before_sha": "${{ github.event.before }}"
      }'
Create a GitHub Actions variable named JATAKA_TEST_MODE and set it based on your need under the actions -> variable.
10
Step 10 - Connect Jira AccountConnect your Jira account so Jataka can read ticket context tied to branches, commits, or pull requests. Ensure the Jira site selected is the one used by your engineering teams.
11
Step 11 - Approve Jira PermissionsAllow all the permissions and click Accept.
12
Step 12 - Set Jira Project KeySet the exact Jira project key that should be used for ticket lookup. Validate the key format and test one linked ticket before enabling for all pipelines.
13
Step 13 - Use Jira Ticket in PR TitleInclude the Jira ticket number in the pull request title so Jataka can derive ticket context for test execution. Example format: PROJ-123: Improve lead conversion validation.
Setup complete: your repository, Salesforce org, secrets, workflow checks, and Jira context are now aligned for automated Jataka validation.

​
MCP Server Setup (VS Code)

To use governed, context-aware coding directly in your IDE, install and authenticate the Jataka VS Code extension using your dashboard auth token. Read the full setup and tool guide here: MCP Server for VS Code