T
Théo JUSTINE
Please add native GitHub actions to HighLevel Workflows so automations can send messages and update repository data without manually building each API request.
Core actions:
  • Create or update a GitHub issue and add a comment.
  • Create or update Markdown/JSON files in a selected repository and branch, with a commit message.
  • Optionally open a pull request for review instead of writing directly to the main branch.
  • Map workflow values and AI-generated output into these actions, and return the resulting GitHub URL/status to the workflow.
My main use case: a scheduled weekly AI report about changes to our HighLevel workflows.
Example prompt: "Every Friday, review all workflows in the selected sub-accounts, explain what changed this week and what stayed unchanged, then publish the report and the supporting data to our GitHub repository."
The report should use actual workflow definitions, version history and audit records: created/deleted workflows, changed triggers, actions, conditions, messages and settings, draft vs published state, who changed what and when where available, plus links/IDs back to the source. Summarize execution results and errors separately from configuration changes.
Provide a readable Markdown report plus structured JSON snapshots/diffs for traceability and later analysis by our team or an external AI assistant. Clearly distinguish verified changes, verified unchanged workflows and unavailable data; missing access must never be reported as "nothing changed".
HighLevel already documents Scheduler, Custom Webhook and Workflow Version History. The request is to connect these capabilities in an easy native GitHub integration and give the reporting AI reliable access to the required workflow history, rather than relying on manually assembled exports.
Allow selection of authorized sub-accounts, repository, branch, folder, schedule and timezone. Respect GitHub permissions/branch protection, keep credentials secure, exclude secrets and unnecessary contact data, and prevent duplicate posts/commits when a run is retried.
This would give agencies a durable, versioned weekly record of their automations and make it easier to understand changes, investigate problems and keep operational documentation up to date.