# AI coding agent metrics

## What is an AI coding agent?

AI coding agents are AI tools responsible for creating pull requests end-to-end — planning work, editing code, running tests, and responding to reviews — not just suggesting code in your editor.

Most often, users assign AI coding agents individually from issue trackers or prompt them to create a single pull request. AI coding agents can also be autonomous, in which case they will have a backlog or objective and continuously pick work items on their own.

### Examples of coding agents

* [**GitHub Copilot coding agent**](https://docs.github.com/en/copilot/how-tos/use-copilot-agents/coding-agent) can create pull requests from [various sources](https://docs.github.com/en/copilot/how-tos/use-copilot-agents/coding-agent), including IDEs, GitHub.com, GitHub issues, GitHub Mobile, CLI, and MCP.
* [**Cursor Cloud Agents**](https://cursor.com/docs/cloud-agent) can create pull requests from [cursor.com/agents](https://cursor.com/agents), or you can trigger them via integrations, such as [Slack](https://cursor.com/docs/integrations/slack) and [Linear](https://cursor.com/docs/integrations/linear).
* [**Claude Code GitHub Actions**](https://code.claude.com/docs/en/github-actions) can create pull requests when you mention `@claude` in PRs or issues.
* [**Claude Code on the web**](https://code.claude.com/docs/en/claude-code-on-the-web) is available in research preview to selected pricing plans and can create pull requests from [claude.ai/code](https://claude.ai/code).

## Understanding your coding agent performance

To help you understand how your AI coding agents are performing, Swarmia automatically detects pull requests created by GitHub Copilot, Cursor, or Claude Code AI coding agents. Swarmia provides a dedicated [**Coding agents**](https://app.swarmia.com/ai/agents/agents) view for analyzing them.

<figure><img src="https://2772466312-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FMa8uBmGhQgR7MTPq9yh7%2Fuploads%2FOLd9tNvSiIjXdNLmekb6%2F003719%402x.png?alt=media&#x26;token=c5522c50-bc0e-4104-a9fc-22fd4d6806fe" alt=""><figcaption></figcaption></figure>

### Agent PR throughput

The Agent PR throughput chart shows how much work AI coding agents help you ship.

**Agent PRs**, **Merged**, and **Closed** show the raw number of AI coding agent pull requests.

**Merge percentage** tells you about your conding agent performance and how effectively people are able to use agents. Ideally, the number should grow close to 100%, but never quite reach it. Even though every closed pull request is technically waste (i.e., work done that was never shipped), a small percentage of closed pull requests indicates that people are experimenting with AI coding agents and pushing their limits.

<figure><img src="https://2772466312-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FMa8uBmGhQgR7MTPq9yh7%2Fuploads%2FX5IwU4RvlErz8Ia35zPQ%2F127.0.0.1_9002_ai_agents_agents_teamId%3D4d9361ab-3217-4690-957e-ca69abe9ca07%26tab%3DPull%2Brequests%26featureGates%3DAiAgentsView%20(1).png?alt=media&#x26;token=995d2ca8-452d-4e3c-bebd-1db2a6283d7f" alt=""><figcaption></figcaption></figure>

### Agent commit percentage

The **Agent commit percentage** chart breaks down your merged pull requests based on how much human intervention was needed for the work to be shipped.

Ideally, all your agent pull requests would have 100% agent commits, which means that the person was able to complete the whole task through the agent interface (e.g., GitHub Copilot's pull request integration). If a pull request ends up in the other buckets, it means that someone needed to pull up their coding environment and commit code to finish the work.

<figure><img src="https://2772466312-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FMa8uBmGhQgR7MTPq9yh7%2Fuploads%2FvHpcP0gnjtIMhp5tXEJq%2F127.0.0.1_9002_ai_agents_agents_teamId%3D4d9361ab-3217-4690-957e-ca69abe9ca07%26featureGates%3DAiAgentsView.png?alt=media&#x26;token=0f841084-737a-48c0-8403-98fe421e7cb4" alt=""><figcaption></figcaption></figure>

### Batch size

The **Batch size** chart helps you compare the distribution of AI coding agent pull requests with your overall [Batch size](https://app.swarmia.com/metrics/code/batch-size) metrics. It also gives you an indication of the complexity of work AI coding agents can tackle.

Depending on how you start using AI coding agents, you might see that most agent pull requests initially fall in the smallest 1–5 or largest 500+ buckets. As your AI coding agent usage develops, you should see the distribution move toward the buckets in the middle. This means your agents can tackle tasks of all sizes and ship code in manageable increments.

<figure><img src="https://2772466312-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FMa8uBmGhQgR7MTPq9yh7%2Fuploads%2F6IZGP6KLhJ4L5Y581wLk%2F127.0.0.1_9002_ai_agents_agents_teamId%3D4d9361ab-3217-4690-957e-ca69abe9ca07%26featureGates%3DAiAgentsView%20(1).png?alt=media&#x26;token=86a17c69-9ff5-4f0f-8665-1e8b19ad86c6" alt=""><figcaption></figcaption></figure>

### Teams breakdown

The "Teams" tab provides a breakdown of the core metrics shown in the top charts, grouped by team.

<figure><img src="https://2772466312-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FMa8uBmGhQgR7MTPq9yh7%2Fuploads%2FTpSzl2JUUfia7Q1om38U%2F127.0.0.1_9002_ai_agents_agents_teamId%3D4d9361ab-3217-4690-957e-ca69abe9ca07%26featureGates%3DAiAgentsView%20(2).png?alt=media&#x26;token=093c59ae-aa59-4da9-b1a5-b403b258e977" alt=""><figcaption></figcaption></figure>

### Pull requests tab

You can use the "Pull requests" tab in combination with the pull request filters at the top to inspect the individual agent pull requests.

<figure><img src="https://2772466312-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FMa8uBmGhQgR7MTPq9yh7%2Fuploads%2FwbtXuDcrSnyuQfOMTxkD%2F127.0.0.1_9002_ai_agents_agents_teamId%3D4d9361ab-3217-4690-957e-ca69abe9ca07%26featureGates%3DAiAgentsView%20(3).png?alt=media&#x26;token=9ecce2bd-ca28-4390-8989-dabd949359bb" alt=""><figcaption></figcaption></figure>

## Identifying PRs created with an AI coding agent

The AI coding agent used to create a pull request is shown on top of the pull request author’s avatar. You can hover over the avatar to see more details.

<figure><img src="https://2772466312-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FMa8uBmGhQgR7MTPq9yh7%2Fuploads%2FwllPIqTPJzh7xE5TczdF%2FScreenshot%202026-01-30%20at%209.17.36.png?alt=media&#x26;token=615a60e6-b6eb-4c3a-a31b-1581c77e2597" alt=""><figcaption></figcaption></figure>

{% hint style="info" %}
In Swarmia, pull requests authored by AI tools are attributed to the people who initiated them. For example, if GitHub Copilot creates a PR on your behalf, Swarmia shows you as its author.
{% endhint %}

## AI agent filter

You can use the **AI agent** filter to view pull requests created by AI agents on all the pages under [Code metrics](https://app.swarmia.com/metrics/code).

(See also the [AI assistant filter](https://help.swarmia.com/use-cases/ai-impact-on-pr-metrics#ai-tool-filters) for viewing pull requests with any kind of AI tools involved. Swarmia considers all AI agent PRs also as AI-assisted.)

<figure><img src="https://2772466312-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FMa8uBmGhQgR7MTPq9yh7%2Fuploads%2FFJO302Zg6vYBklBrBzg9%2FScreenshot%202026-01-30%20at%209.03.16.png?alt=media&#x26;token=5e6d178a-0f6e-42e2-8bf3-15d5447b7218" alt=""><figcaption></figcaption></figure>

## Frequently asked questions

### Which AI coding agents are supported?

We currently support GitHub Copilot, Cursor, and Claude Code. ([See examples above](#examples-of-coding-agents).) If you're using a different coding agent or have built your own custom AI coding agents, reach out to us at <hello@swarmia.com> with the agent details (the agent's GitHub account or Git commit email, plus an example pull request) you'd like to track.

### Why is my PR not attributed to an AI coding agent?

Swarmia attributes a PR to a coding agent if the PR or its first commit has been authored by an AI agent.

Example 1: With [GitHub Copilot coding agent](https://docs.github.com/en/copilot/how-tos/use-copilot-agents/coding-agent), GitHub shows *Copilot* as the PR author. Swarmia shows this as a GitHub Copilot agent PR:

<figure><img src="https://2772466312-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FMa8uBmGhQgR7MTPq9yh7%2Fuploads%2FTWx9fyT3ujXAYKK0qXnE%2F003555%402x.png?alt=media&#x26;token=15aca564-cdd7-4bff-8329-1391f3b6d2b4" alt=""><figcaption></figcaption></figure>

{% hint style="info" %}
Note: Regardless of what GitHub displays, Swarmia always shows the author as the person who initiated the AI coding agent PR. [Identifying PRs created with an AI coding agent](#identifying-prs-created-with-an-ai-coding-agent) above for more details
{% endhint %}

Example 2: When using [Cursor Cloud Agents](https://cursor.com/docs/cloud-agent), GitHub shows *cursoragent* as the author of the first commit, and the user who started the agent as the PR author. Swarmia shows this as a Cursor agent PR:

<figure><img src="https://2772466312-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FMa8uBmGhQgR7MTPq9yh7%2Fuploads%2FPaIEEcLnG09Sx0IrUGZa%2F003554%402x.png?alt=media&#x26;token=c9eb6b9d-7cb6-47a0-a58a-9215d3bbee70" alt=""><figcaption></figcaption></figure>

Example 3: The first commit in this PR was authored by *Claude* but committed by *cursoragent*. Swarmia shows this as a Claude Code agent PR:

<figure><img src="https://2772466312-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FMa8uBmGhQgR7MTPq9yh7%2Fuploads%2Fcw8CMcSjcmJP4maY2PqE%2F003647%402x.png?alt=media&#x26;token=619a1644-bd84-4ab3-b2b7-0ebcec590f03" alt=""><figcaption></figcaption></figure>

### If I ask an AI tool to create a PR with my local changes, is it considered an AI coding agent PR?

No, we don't regard that as an AI coding agent PR since you review and accept the suggestions (and perhaps manually write some of the code) before creating the PR.

The AI uses your local Git identity, so you are marked as the PR author and the commit author. (The tool might mark itself as the co-author, however.)

Swarmia [shows these PRs as AI-assisted](https://help.swarmia.com/use-cases/ai-impact-on-pr-metrics#automatic-ai-assistant-detection-for-prs).

### What's the difference between the "AI assistant" and "AI agent" filters?

* **AI assistant:** filters pull requests assisted by AI tools in general, and can be detected with heuristics like recent AI-tool usage. See the [AI assistant filter](https://help.swarmia.com/use-cases/ai-impact-on-pr-metrics#ai-tool-filters).
* **AI agent:** [filters pull requests](#ai-agent-filter) created by an AI tool end-to-end. AI agent PRs are also counted as AI-assisted. See [examples of AI coding agents](#examples-of-coding-agents) above.

## Further reading

Swarmia blog: [Five levels of AI coding agent autonomy, and why higher isn’t always better](https://www.swarmia.com/blog/five-levels-ai-agent-autonomy/) by Miikka Holkeri, Product Manager · Mar 19, 2026