Custom AI tool integrations

Link contributors or push usage data to track any custom AI coding tool in Swarmia.

Read more about Swarmia's other AI coding tool integrations and understanding the impact of AI tools.

Swarmia has built-in integrations for GitHub Copilot, Cursor, Claude Code, Codex, and Devin. It also detects the built-in review agents, such as CodeRabbit, automatically. For any other AI coding tool, you can create a custom integration, so it shows up alongside the built-in ones in AI adoption and AI tool detection and filters.

How it works

There are two ways to connect a custom AI tool to your data, and you can use either or both for the same tool: link contributors or push usage data.

Add a custom AI tool

You need the organization admin role to manage integrations.

1. Go to Settings → AI tools and find the Other AI tools section.

2. Click Add new.

3. Enter a Name (shown in reports) and an API identifier (aiService in the additional AI integrations API). The identifier can't be changed after creation.

4. Click Save.

The tool's row shows when Swarmia last received data for it.

To rename a tool or delete it, click the edit icon on its row. Deleting a tool also removes all of its usage data.

Link contributors

Linking contributors is optional. Leave Linked contributors empty when you only want to send usage data.

If your custom tool commits, opens pull requests, or leaves reviews under its own Git identity, you can link that contributor to the tool. Swarmia then attributes any commit, co-authored commit, or review from the linked contributor to the tool with high confidence, the same way it recognizes built-in tools like GitHub Copilot or Claude Code. This allows the tool to show up as a cloud agent or review agent.

To link a contributor:

1. Go to Settings → AI tools and find the Other AI tools section.

2. Click the edit icon on the tool's row (or Add new to create one first).

3. Under Linked contributors, search for a contributor(s) by their GitHub or GitLab username and select them.

4. Click Save.

A contributor can only be linked to one custom AI tool at a time. If you try to link a contributor who's already linked elsewhere, they show as unavailable in the search.

Send usage data to API

Sending usage data to the API is optional. Use it to track adoption and seats for a custom AI tool.

For tools without their own Git identity, or when you want to track adoption or seats, you can push daily per-user usage data to Swarmia from your own scripts or systems. For each user and day, report whether they had access to the tool (is_enabled) and whether they actively used it (is_active) — that's what separates active seats from idle ones in AI adoption.

You can also send optional details like message counts, subscription tiers, spend, and breakdowns by model, client, or programming language. Swarmia stores this detail, but it isn't visible in the app yet.

Swarmia matches each record to an author by email. It stores the usage as daily metrics, and sending the same user, tool, and date again overwrites the previous values, so you can safely re-send a day's data.

Because the data is daily and not tied to individual commits, pull requests are attributed to the tool using time-based (low-confidence) AI-assisted attribution: a pull request counts as assisted when its commit author used the tool within the previous 24 hours.

Usage data is submitted to the additional AI integrations API, which authenticates with an API token. Generate one with the AI Usage scope in Settings → API tokens. The API reference documents the endpoint, the full request schema, and every field you can report.

You don't need to add the tool in advance — posting data for a new identifier creates it automatically.

Frequently asked questions

How is this different from the built-in integrations?

The GitHub Copilot, Cursor, Claude Code, and Codex integrations fetch data for you once you connect them. Devin and review agents such as CodeRabbit are detected from their Git activity without any setup, but Swarmia doesn't fetch their usage data. With a custom AI tool, you're responsible for either linking its contributors or collecting and pushing its usage data yourself.

Where does the data show up?

If you link a contributor account to the tool, the data shows up in AI Impact, Cloud agents and Review agents pages as high confidence attributions.

If you push your usage data, the data shows up in AI adoption, alongside the built-in tools, once the tool has received user data. The tool also appears in AI tool detection and filters, where pull requests are attributed to it through time-based (low-confidence) AI-assisted attribution.

Spend data for custom AI tools is not visible yet. If you are interested in seeing it, please contact [e-mail].

Can I backfill historical data?

Yes, you can send data for any dates via the additional AI integrations API.