Setup Deployment Agent
Configure integrations
Make sure to have configured your GitHub, GitLab, Azure Pipelines or Bitbucket integration so Deployment Agent can post its help in Pull Request comments.
If you want to add AI to Deployment Agent, follow sfdx-hardis AI setup.
You can also receive Slack, Microsoft Teams and Email notifications in case of successful deployment.
If you configure Jira or Generic Ticketing integrations, ticket numbers will be extracted and displayed in the Pull Request comment.
If you want to automatically fix deployment errors using coding agents (Claude, Codex, Gemini, Copilot), see Coding Agent Auto-Fix setup.
Using sfdx-hardis CI/CD
If you are using sfdx-hardis CI/CD, you are already all set.
Using custom CI/CD pipeline
Replace your calls to Salesforce CLI commands with calls to the sfdx-hardis wrapper commands.
Configure your GitHub, GitLab, Azure Pipelines or Bitbucket integration so Deployment Agent can post its help in Pull Request comments.
Notes:
- sfdx-hardis Deployment Agent works better with the --json option, please use it.
Example
Replace:
sf project:deploy:start -x manifest/package.xml --checkonly
with:
sf hardis:project:deploy:start -x manifest/package.xml --checkonly
Advanced example
Replace:
sf project deploy start --dry-run --source-dir force-app --ignore-warnings --ignore-conflicts --test-level RunLocalTests --coverage-formatters json-summary --verbose --wait 120 --json
with:
sf hardis:project:deploy:start --dry-run --source-dir force-app --ignore-warnings --ignore-conflicts --test-level RunLocalTests --coverage-formatters json-summary --verbose --wait 120 --json
Not updating custom CI/CD pipeline
You don't want to update your calls to sf project deploy start?
That's fine: you won't benefit from the error management, but you can still benefit from the Flow Visual Git Diff and the other integrations.
Just add the Notify command to your custom CI/CD pipeline.