[fix] Extra changes related to corrupt SDK evals

[junaway]

No description provided.

[vercel]

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

Project	Deployment	Actions	Updated (UTC)
agenta-documentation	Ready	Preview, Comment	Feb 20, 2026 3:03pm

[Copilot]

Pull request overview

This PR fixes issues related to corrupt SDK evaluations by improving robustness in application name handling and evaluation slug management.

Changes:

• Replaces unreliable ApplicationRevision.name with authoritative SimpleApplication.name by fetching from the API
• Removes manual slug computation in favor of using the application_revision.slug attribute directly
• Adds defensive name sanitization for evaluation runs to handle None, empty, or whitespace-only names

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 1 comment.

File	Description
sdk/agenta/sdk/managers/applications.py	Adds _fetch_simple_application() function to retrieve authoritative application names and updates name preservation logic to use it instead of ApplicationRevision.name
sdk/agenta/sdk/evaluations/preview/evaluate.py	Removes get_slug_from_name_and_id import/usage in favor of using application_revision.slug directly, adds name sanitization with "SDK Eval" fallback, and increases evaluator trace retry count from 20 to 30 for consistency

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[bekossy]

@codex review

[chatgpt-codex-connector]

Codex Review: Didn't find any major issues. Swish!

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".