[fix] Align demo agent defaults

[mmabrouk]

Context

In the demo cloud deployment, Daytona is the only enabled execution environment, but a new agent still loads the unavailable local sandbox default. Users must open the execution settings and select Daytona before the agent can run. New agents also start on GPT-5.5 and expose the opinionated Pi (Agenta) harness even though that harness should not be selectable here.

Changes

When the current sandbox value is missing from the enabled deployment options, the frontend now selects the first available option. A Daytona-only deployment therefore changes from an invalid local selection to Daytona automatically, while deployments that offer the configured default keep it unchanged.

The shared agent template, service fallback, and runner file now default to openai/gpt-5.6-luna. The harness picker filters pi_agenta from both the capability-backed rail and the schema-backed fallback picker. Plain Pi remains the default harness.

How to review

Start with useModelHarness.tsx and HarnessSelectControl.tsx for the unavailable-sandbox fallback and harness visibility rules. Then check types.py, the service config, and agent.json for the synchronized model default. The service test pins the published SDK and inspect defaults to the same model.

Tests / notes

• pnpm lint-fix
• pnpm --filter @agenta/entity-ui run types:check
• pnpm --filter @agenta/entity-ui test:unit -- tests/unit/enumSelectControl.test.ts tests/unit/connectionUtils.test.ts (201 passed)
• Focused service tests for the default template and config fallback (9 passed)
• Ruff format and check on the touched Python files

What to QA

• Open a new agent in the Daytona-only demo deployment. Execution environment shows Daytona without a manual selection.
• Create a new agent. Its model is GPT-5.6 Luna and plain Pi remains selected.
• Open the harness picker. Pi (Agenta) is absent, while plain Pi and Claude Code remain available.

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
agenta-documentation	Ready	Preview, Comment	Jul 14, 2026 1:44pm

[coderabbitai]

Important

Review skipped

Auto reviews are disabled on base/target branches other than the default branch.

Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Pro Plus

Run ID: 9b394ec3-d438-43e6-ba29-8673cad3cbfb

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch fix/agent-demo-defaults

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[mmabrouk]

🤖 The AI agent says: Please review the unavailable-sandbox fallback and confirm that hiding pi_agenta while keeping plain Pi as the default matches the intended demo-cloud product behavior.

[github-actions]

Railway Preview Environment

Preview URL	https://gateway-production-120f.up.railway.app/w
Project	agenta-oss-pr-5311
Image tag	pr-5311-28a4e80
Status	Deployed
Railway logs	Open logs
Workflow logs	View workflow run
Updated at 2026-07-14T14:19:31.305Z