Sign in
← AI PulseJul 24, 2026

Wire · news · Single-source brief

GitHub Issues Introduces Agent Automation Controls in Public Preview

GitHub Issues has launched new controls for agent automations in public preview, allowing users to review changes, understand their rationale, and configure confidence thresholds before application.

By Illumora Editorial

Source · Jul 23, 2026, 3:30 PM · On Illumora · Jul 24, 2026, 2:12 AM

Rewritten from one allowlisted primary — not independent enterprise reporting. Lanes →

Brief drafted by Illumora’s editorial model from the linked primary source. Ops desk reviews flagged pieces. How we write →

Read the source →GitHub Changelog (Copilot) — Agent automation controls in GitHub Issues in public preview - GitHub Changelog
Save

GitHub has released new agent automation controls for GitHub Issues in public preview. These controls enable users to manage how agent automations label, type, assign, and close issues. The update provides visibility into the rationale behind each change and offers a review process before changes are applied.

Key Points

  • GitHub Issues now displays the reason for each automated change.
  • Users can review automated changes before they are applied.
  • A new search filter, has:suggestions, helps find issues with pending changes.
  • Repository administrators can configure automation levels by setting a confidence threshold.
  • Rationale, confidence, and approvals work with GitHub Agentic Workflows and Copilot cloud agent automations.
  • These capabilities are also accessible via the REST and GraphQL APIs.
  • At launch, the controls cover changes to labels, fields, type, close, and assignees.

Context

According to GitHub, agent automations increasingly handle tasks such as labeling, typing, assigning, and closing issues. The new controls aim to balance automation efficiency with user oversight. The system allows users to decide the extent of automation, providing a review step for changes that might require human intervention.

Why It Matters

These new controls offer developers and repository managers greater flexibility in integrating AI-powered automation into their workflows. The ability to review changes and understand their rationale can help maintain accuracy and control, particularly in complex or public repositories, while still benefiting from automated efficiency.

What To Do

  • Create an automation from the Automations pane in your repository's Agents tab to explore the new features.
  • Use the has:suggestions filter in issue search to locate issues awaiting review.
  • Repository administrators should configure automation levels by setting confidence thresholds.
  • For existing GitHub Agentic Workflows, upgrade workflows to add issue intent support.
  • Set issue-intents: true in workflow frontmatter to require intents for specific safe outputs.
  • Set issue-intents: false to opt out of issue intents.

Keep Exploring

/atlas/claude-family /atlas/gpt-family /atlas/**gemini**-family /atlas/llama-open