Monitor. Govern. Scale.

Central Agentic Ops

Monitor and control GitHub Agentic Workflows at scale.

Keep track of all runs, costs, and outcomes of your GitHub Agentic Workflows across repositories. Set execution scope and rollout policy.

Explore CAO Get started Sample Dashboard

Central Agentic Ops mobile dashboard showing weekly workflow rhythm, workflow health, repository coverage, retained issues and pull requests, and observed grader results. — Central Agentic Ops mobile dashboard overview

One workflow ecosystem

Shared work. Clear ownership.

Enterprise teams, organizations, and repository maintainers own their workflows. Shared work reaches enrolled repositories alongside local automation, with explicit scope and ownership.

Enterprise

Centrally managed

Optimization

Org 1

Centrally managed

Issue triage

Repo 1

Repo-managed · monitored

Daily team status

StatusOptimizeTriage

Repo 2

Repo-managed · monitored

Test coverage

TestsOptimizeTriage

Org 2

Centrally managed

Release readiness

Repo 3

Repo-managed · monitored

Release checks

ChecksOptimizeReadiness

... more organizations... more repositories

Org N

Centrally managed

Organization standards

Repo N

Repo-managed · monitored

Project health

HealthOptimizeStandardize

... Org N / Repo N

Setup, authentication, and rollout

Roll out with confidence.

Bring an existing workflow, such as Daily Repo Status from Agentics, under central control. Set repository scope and separate read/write credentials. Review proposed reports before authorizing live delivery. Keep rollout policy and steering in your control repository, and reuse the workflow across enrolled repositories. Add campaigns when you need coordinated work.

$curl -fsSL https://raw.githubusercontent.com/githubnext/gh-cao/main/install.sh | bash

==> CAO runtime materialized and verified.

$gh cao setup

Next: gh cao doctor
Then review git diff and git status, commit, and push the installed files.

Example agent request — not a CLI command

Adapt Daily Repo Status for CAO

Import shared/control.md as standalone.
Bind reads and review outputs to admitted targets.
Keep policy disabled, review-only, and scoped to one repo.
Preserve upstream ownership. Do not enable or run.

Local workflows, central visibility

Monitor what runs.

See workflow health, cost, and outcomes in one dashboard. Find what needs attention, then drill into the run and its evidence.

Mobile CAO Runs view showing its compact app header and stacked workflow run chart by success, failure, cancellation, and unknown conclusion.

Earn trust before expanding autonomy

Roll out safely.

Review proposed actions before enabling live changes through explicit policy. Expand scope with evidence, not just successful runs: execution success is not an accepted outcome.

Mobile Operational Value view for githubnext/gh-cao showing AI credit per successful optimization-token-optimizer run over time.