Skip to main content
Every application decides what Ankra AI does when its repository gets a pull request. Open Applications → <your application> → Settings → Who runs it (/next/applications/<application-id>?view=settings&section=ai); the On your pull requests section lists each activity. Each activity follows your organisation’s settings until you change it here. The section is a view onto the same per-repository settings the reviews read when a pull request event arrives, so what you set here and what your organisation’s source-control connection sets can never disagree.

Before you start

  • A role that may change the application. Without it, Configure is greyed out and the section reads Your role can’t change what Ankra does on pull requests.
  • A repository connected through the Ankra GitHub App. Without one, the section says why instead of listing activities: the application has no repository, the App is not installed for the repository’s owner, or the repository is on a provider these activities cannot use. Install the App under Ankra AI → Gateway connections → GitHub → Install the GitHub App. See GitHub.

The activities

Each row reads On or Off, then Follows your organisation or Set for this application, and Your organisation has turned this off when the organisation’s switch is off. Every activity is off for a repository nobody has configured. A fresh source-control connection never starts reviewing, replying or deploying on its own, so each activity is switched on deliberately - here, or for the whole connection under AI code review. The platform also lists a code review of pushed branches and a review of failed workflow runs. Neither runs, so the section does not show them.

Change an activity

1

Open its settings

Press Configure on the activity’s row.
2

Choose where its settings come from

Under the settings source, pick Follow organisation defaults or Customise for this application.With Follow organisation defaults, the dialog says This activity uses the organisation’s current settings, including future changes to those defaults.
3

Set its values

With Customise for this application, tick Enable this activity for the application and set the values in What you can set. If the organisation has turned the activity off, the dialog warns that saving an override will not enable it globally.
4

Review and save

Press Review AI settings. The Save AI activity step sums up the change, for example the model and the limits. Confirm it.The section shows Application AI settings saved., and the row now reads Set for this application.
Customising an activity pins the whole activity: the values in the dialog when you save become this repository’s settings for it, and later changes to the organisation default stop reaching it. To undo every customisation at once, press Use organisation defaults for all activities and confirm Restore organisation AI defaults. The button appears only while some activity is set for this application.

What you can set

The pull request review reports findings. It does not propose fixes on its own. Preview URLs has no settings of its own here - it is on or off. What a preview deploys comes from the application’s demo defaults; see environment variables and a throwaway database. Turning it off stops the automatic pull-request previews for this repository without touching manually launched preview demos.
Mentioning @ankraai in a pull request comment gets an answer whenever Replies to mentions is on for the repository: a mention reply is a direct request from a person, so it is never filtered by branch scope, draft state, or either review budget. The run still counts toward the budgets.

Which skills your organisation’s review rules use

Your organisation’s review rules is a second pass that rides on the pull request review and appends an Organisation skills section to the same comment, so it runs only when Pull request review runs too. It judges the change against:
  • every skill marked Always in use: Ankra AI reads it in every conversation, and
  • the skills pinned in the activity’s skill_ids.
Skills of every kind count: core, CI/CD and add-on alike. One review uses at most 12 skills, always-active ones first, with at most 24,000 characters of skill text. With no such skill, it posts nothing. Only this activity reads skill_ids (the skills lane in ankra application ai-config). The field also appears on the other activities in the CLI and API output, but no review reads it there. The portal has no skill picker: to put a skill in front of the review there, mark it Always in use, which also adds it to every Ankra AI conversation, or pin it with the CLI. See AI Skills. To steer the pull request review itself, use its review instructions: per connection under AI code review, or per repository as Review instructions on the pull request review (custom_instructions in the CLI). Customising it pins the whole activity, as described under Change an activity.

Settings are per repository, not per application

These settings are keyed by the repository, not by the application. Two applications connected to the same repository share one set of settings - turning an activity off on one turns it off for the other, and each application’s section shows the other’s edits as its own. Preview URLs also fires once per matching application, so such a repository collects one preview comment per application on every pull request. Keep one application per repository.