Skip to content

Add agent skill to proactively identify failing tests#1545

Merged
nathanjmcdougall merged 2 commits into
mainfrom
copilot/add-agent-skill-identify-failing-tests
Mar 25, 2026
Merged

Add agent skill to proactively identify failing tests#1545
nathanjmcdougall merged 2 commits into
mainfrom
copilot/add-agent-skill-identify-failing-tests

Conversation

Copilot AI commented Mar 25, 2026

Copy link
Copy Markdown
Contributor

In #1527, the agent changed rule management logic but didn't run test_readme_example tests, which assert exact CLI output including rule selections. CI caught it instead.

New skill usethis-python-test-affected-find teaches agents to identify indirectly affected tests before pushing:

  • General principles: trace callers not just callees, check integration tests, flag tests with exact output assertions
  • Domain-specific rules: rule management changes → always run test_readme_example; tool config changes → run CLI output tests; file I/O changes → run all dependent tool tests
Original prompt

This section details on the original issue you should resolve

<issue_title>Agent skill to proactively identify failing tests</issue_title>
<issue_description>In #1527, relevant affected tests weren't identified by the agent, so the CI failed.

We should try and create an agent skill usethis-python-test-affected-find which explains how to find potentially failing tests, giving some general principles, as well as specific guidance. The first piece of specific guidance is that for any changes to rule management logic, we should always run the test_readme_example tests - these are always potentially affected.</issue_description>

Comments on the Issue (you are @copilot in this section)


⚡ Quickly spin up Copilot coding agent tasks from anywhere on your macOS or Windows machine with Raycast.

Create a new agent skill that provides guidance for proactively
identifying tests that are potentially affected by code changes,
to catch regressions before CI.

Includes general principles (trace callers, check integration tests,
check exact-output tests) and domain-specific rules (rule management
changes should always trigger test_readme_example tests).

Closes #1528

Co-authored-by: nathanjmcdougall <18602289+nathanjmcdougall@users.noreply.github.com>
Agent-Logs-Url: https://github.com/usethis-python/usethis-python/sessions/3215728e-fb96-40f4-b950-1616594dbc98
Copilot AI changed the title [WIP] Add agent skill to proactively identify failing tests Add agent skill to proactively identify failing tests Mar 25, 2026
Copilot AI requested a review from nathanjmcdougall March 25, 2026 20:18
@nathanjmcdougall nathanjmcdougall marked this pull request as ready for review March 25, 2026 20:30
@nathanjmcdougall nathanjmcdougall merged commit abdc820 into main Mar 25, 2026
@nathanjmcdougall nathanjmcdougall deleted the copilot/add-agent-skill-identify-failing-tests branch March 25, 2026 20:30
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.

Agent skill to proactively identify failing tests

2 participants