Integrated AI

Pragma's own AI is small on purpose: it writes the commit, drafts the pull request, edits the file under your cursor, and answers the quick question. It is not another agent — your agents do the heavy lifting.

Setup

The first launch shows Enable AI features: connect a provider by subscription sign-in (browser OAuth) or by pasting an API key. Featured providers:

• Claude Code (Claude Pro/Max)
• Codex (ChatGPT Plus/Pro)
• GitHub Copilot
• OpenRouter
• Google AI (Gemini)
• opencode Go

More options lists the rest (Anthropic, OpenAI, and others). Connect as many as you like; Add another provider chains them. Skip is always available, and Settings → AI lists connected accounts with sign-out.

Your credentials, your accounts, your usage — nothing is proxied through a Pragma service.

The same step offers an optional System 1 model card at the top. That key powers Auto mode in the agent pickers and, together with these helpers, AI merge-conflict resolution.

Inline edit

In any code editor, select lines and press ⌘KCtrl+K:

1. A pill opens under the selection; describe the change.
2. The result lands as red/green hunks with accept/reject bars.
3. Accept hunks one at a time (⌘↵Ctrl+↵, or +Shift for all), reject with ⌘⌫Ctrl+Backspace, walk hunks with Alt+↑/↓, or Abort while running.

The model gets read-only tools — read, grep, find, list — so it understands context, but nothing touches disk until you save. Inline edit is available on local worktrees, not over SSH.

Commit messages and pull requests

• The Changes tab's commit box: Shift+Tab generates a commit message from the staged diff.
• The PR create view: Shift+Tab drafts a title and body from the branch diff.
• The Commit & PR button (right sidebar header) plans and makes all the commits, pushes, and drafts the PR in one action.
• Commit and Push Fixes on a pull request's merge card tells the model the commit resolves merge conflicts, so the message says so.
• The agent board's "Commit all and open PR" and a fanout pick's final commit message use the same pipeline.

Ask from the palette

Type a question in the command palette and the top row becomes Ask AI {message}: a streamed answer scoped to the current project and worktree, with read-only tools. Escape cancels. Not available over SSH.

System 1 models (Jev)

The helpers above are chat models. A System 1 model is a fast classifier that answers typed questions with probabilities instead of text, in well under a second. Pragma uses TypeSafe's Jev by default, through TypeSafe or OpenRouter, and any Jev-compatible endpoint works. Connect one in Settings → AI (setup steps are in Auto mode) and it powers:

• Auto mode — picks the agent, model, and reasoning effort for a launch from your prompt, live harness and model benchmarks, and automode.md.
• Agent progress — each running agent's line in the sidebar names what it is doing (planning, coding, testing, …) beside a progress estimate.
• Merge-conflict resolution — picks ours, theirs, or both for every conflict, then hands the uncertain files to the built-in AI above to verify.

The key is stored in an owner-only file and is only sent to the endpoint you set.