Skip to content

Improve agentic maintenance and dotnet-test evaluations - #4

Open
Evangelink wants to merge 6 commits into
mainfrom
fix-stale-pr-cleanup
Open

Improve agentic maintenance and dotnet-test evaluations#4
Evangelink wants to merge 6 commits into
mainfrom
fix-stale-pr-cleanup

Conversation

@Evangelink

Copy link
Copy Markdown
Owner

Why

Agentic maintenance needed more reliable stale-PR pagination, while several dotnet-test skills and evaluations were carrying unnecessary cost or lacked outcome-focused coverage. Issue dotnet#900 also identified one migration skill to trim and another to strengthen.

What changed

  • Upgrade the agentic workflow assets and paginate stale pull request cleanup, including the follow-up workflow review fixes.
  • Streamline code-testing agent guidance and add focused evaluation coverage for source/test pairing and grading behavior.
  • Reduce the always-loaded xUnit-to-MSTest migration skill by about 80%, keeping decision-critical safety rules inline and moving conditional runner mappings to its existing reference.
  • Make MSTest v1/v2-to-v3 migrations evidence-driven and executable, including compatible Microsoft.NET.Test.Sdk alignment and build/test parity checks.
  • Add a net8 migration trial that verifies package updates, DataRow correction, and actual passing tests in both evaluation formats.

The MSTest v3-to-v4 no-verdict item from dotnet#900 is intentionally unchanged because dotnet#909 established it as judge infrastructure rather than a skill-content defect.

Validation

  • Skill Validator checks pass for both modified migration skills.
  • Both migration evaluation YAML files parse successfully.
  • The migrated net8 fixture passes all 5 tests.

github-actions Bot and others added 6 commits June 22, 2026 14:41
Upgrade the generated agentic workflow assets and ensure stale PR discovery includes every result page and draft pull requests.

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

Copilot-Session: 968a22c2-327f-4d26-8f86-1c59bcd323ea
Address remaining high-confidence items from dotnet#899 by bounding the code-testing pipeline and adding eval coverage for grade-tests and find-untested-sources.

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

Copilot-Session: e430fee9-d3df-4ef5-85a4-745ae4b17046
Pin the Copilot setup checkout action and include the cutoff date in stale PR search results.

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

Copilot-Session: 968a22c2-327f-4d26-8f86-1c59bcd323ea
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

Copilot-Session: 5e19d263-02a6-45b1-9cfb-424fa4d10863
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant