Various integrations and resources need to be set up for Mayil to be fully functional. All configurations are made through Hashicorp Vault, a secret management tool running on the deployment we just installed.

​
Using Vault

Vault should run on the base domain we deployed Mayil to in the previous step (such as mayil.companyname.com). When opened for the first time, you should see a login page like this-

We will use the username method to login. Use envVars.CLIENT_ID and envVars.CLIENT_SECRET provided in values.yaml in the helm installation package you received from us.

Please update this default username and password after logging in. Vault supports most popular methods including GitHub, LDAP, AppRole, and more. See here.

Once logged in, under Secrets Engine, open mayil ➤ data ➤ user_config. You should see something like this-

This file contains all the configuation options for Mayil. When installing integrations, this file will have to be modified with details about that integration. Every time this file gets updated, Mayil will refresh.

​
Verifying config values

Go to <deployment>/api/health (such as `https://mayil.companyname.com/api/health) to see the status of the configuration. Any errors in the config will also be listed there. When config is valid, you will see something like-

​
Sample configuration

Here is a sample config file-

{
  "data": {
    "ENABLED": true,
    "ENDPOINTS": [
      {
        "deployment": "4Turbo1106",
        "deployment_type": "azure_oai",
        "endpoint": "https://mayil-ai.openai.azure.com/",
        "key": "GPT4",
        "token": "<TOKEN>"
      },
      {
        "deployment": "35Turbo1106",
        "deployment_type": "azure_oai",
        "endpoint": "https://mayil-ai.openai.azure.com/",
        "key": "GPT3",
        "token": "<TOKEN>"
      },
      {
        "deployment": "GPT4O",
        "deployment_type": "azure_oai",
        "endpoint": "https://mayilus.openai.azure.com/",
        "key": "GPT4O",
        "token": "<TOKEN>"
      },
      {
        "deployment": "embedding",
        "deployment_type": "azure_oai",
        "endpoint": "https://mayil-ai.openai.azure.com/",
        "key": "EMBEDDING",
        "token": "<TOKEN>"
      }
    ],
    "EXTERNAL_DOCUMENTATION_SOURCES": [""],
    "GIT_REPOS": [
      {
        "git_url": "https://github.com/ORGNAME/REPONAME",
        "repo_name": "ORGNAME/REPONAME",
        "repo_summary": "Small python program to calculate the area of different shapes",
        "tags": [],
        "tracker_ids": [
          "My Kanban Project"
        ]
      }
    ],
    "INTEGRATIONS": [
      {
        "integration_type": "github",
        "identifier": "962458",
        "organization_id": "Mayil-AI",
        "secret": "<BASE64_ENCODED_PRIVATE_KEY",
        "url": "https://github.com",
        "valid_ids": [
          "11111111",
          "22222222",
        ],
        "webhook_secret": "<CUSTOM SECRET>"
      },
      {
        "integration_type": "jira",
        "identifier": "Q1oS2hfmQugWAOJ2rxZr208xeMumFTwZ",
        "ignore_labels": [
          "mayil_norun",
          "test_issue"
        ],
        "project": "My Kanban Project",
        "secret": "<JIRA SECRET>",
        "url": "https://mayiltest.atlassian.net",
        "webhook_secret": "<WEBHOOK SECRET>"
      }
    ],
    "TELEMETRY_LEVEL": "standard"
  },
  "options": {},
}

​
Configuration options

  1. ENDPOINTS- See OpenAI

  2. EXTERNAL_DOCUMENTATION_SOURCES takes a list of strings, each pointing to a documentation source that could be relevant to Mayil. For instance, if you are setting up Mayil to work for a project that is a Java Spring app, then the Spring documentation would be relevant. The more relevant documentation Mayil has access to, the more knowledgeable it becomes.

    For now, EXTERNAL_DOCUMENTATION_SOURCES should be a subset of ["Shopify", "eslint", "spring"]. Please leave blank if there are no relevant sources.

    These documentation sources need to be set up by the Mayil team. Please reach out to us if you would like a new source.

  3. TELEMETRY_LEVEL can be one of [ "minimal", "standard", "verbose"].

  4. ENABLED can be one of ["true" or "false"]. Set to false to run Mayil with default values.

HelmGitLab