Skip to content
Merged
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
42 changes: 33 additions & 9 deletions README.md
Original file line number Diff line number Diff line change
@@ -1,24 +1,48 @@
# agentic-ops
# agentic-ops — audit token spend, surface waste, and optimize agentic workflows with confidence

[![CI](https://github.com/githubnext/agentic-ops/actions/workflows/ci.yml/badge.svg)](https://github.com/githubnext/agentic-ops/actions/workflows/ci.yml)

This repo contains a small set of GitHub Agentic Workflows for auditing Copilot token usage and highlighting workflows that should be optimized.
`agentic-ops` is a focused bundle of GitHub Agentic Workflows for teams scaling agentic automation and wanting better visibility into cost, usage, and optimization opportunities. Instead of guessing which workflows are expensive or where token waste is hiding, this package gives you an audit trail, historical reporting, and conservative recommendations you can review before making changes.

## Usage
## Introduction

To add one of these workflows to your repo, use `gh aw add <owner>/<repo>/<workflow-name>`.
It is built for platform engineers, developer productivity teams, and repository maintainers who are scaling agentic workflows and need a practical way to keep them efficient. The bundle helps solve a common problem with AI automation: token usage grows quickly, but the signals for where to improve are scattered across workflow runs and logs. With `agentic-ops`, you get repeatable workflows that make usage measurable, optimization opportunities actionable, and efficiency work easier to operationalize.

## Key Features

- **Clear operational visibility** with a daily token audit that captures usage, cost, trends, and workflow-level hotspots.
- **Actionable optimization guidance** that identifies high-token workflows and proposes safe, conservative improvements.
- **Faster cost control** by helping teams find waste before it becomes a recurring operational expense.
- **Built for real GitHub workflows** using GitHub Agentic Workflows, so installation and adoption fit naturally into existing repositories.
- **Useful historical context** through shared snapshots that support trend analysis instead of one-off debugging.
- **A focused bundle** that gives you both measurement and optimization, not just another standalone report.

## Quick Start

Install the bundle with `gh aw add`:

```bash
gh aw add githubnext/agentic-ops/copilot-token-audit githubnext/agentic-ops/copilot-token-optimizer

# Then compile the installed workflows in your repository
gh aw compile
```

This adds the workflow to `.github/workflows/`. For guided setup, use `gh aw add-wizard githubnext/agentic-ops/copilot-token-audit`.
After installation, you can use the included workflows to:

Release history lives in [CHANGELOG.md](CHANGELOG.md).
- run a daily audit of workflow token usage
- identify the workflows consuming the most tokens
- generate optimization recommendations grounded in recent run data

## Workflows
Included workflows:

| Workflow | What it does |
| ----- | --- |
| [`Daily Copilot Token Usage Audit`](https://github.com/githubnext/agentic-ops/blob/main/workflows/copilot-token-audit.md?plain=1) | Collects recent Copilot workflow usage and creates a daily audit snapshot. |
| [`Copilot Token Usage Optimizer`](https://github.com/githubnext/agentic-ops/blob/main/workflows/copilot-token-optimizer.md?plain=1) | Analyzes expensive workflows and proposes conservative token-reduction changes. |
| [`Daily Token Usage Audit`](https://github.com/githubnext/agentic-ops/blob/main/workflows/copilot-token-audit.md?plain=1) | Collects recent workflow token usage, stores historical snapshots, and publishes a daily audit summary. |
| [`Token Usage Optimizer`](https://github.com/githubnext/agentic-ops/blob/main/workflows/copilot-token-optimizer.md?plain=1) | Analyzes high-token workflows and recommends conservative efficiency improvements backed by recent run data. |

Release history lives in [CHANGELOG.md](CHANGELOG.md).

## License

MIT