From f4a0a8dba8fbd42be63760b5cb6fe481be578b7b Mon Sep 17 00:00:00 2001 From: Andrew Nester Date: Fri, 6 Nov 2026 11:51:59 +0100 Subject: [PATCH] pr-review skill: ask whether API changes were run on a real workspace Pass 9 of the pr-review skill reviews tests but never asks whether a change that sends new values to a Databricks API was verified against a real workspace. The fake server in libs/testserver only returns what its handlers were taught, so a test that records the request shows what the CLI sends, not what the API accepts. Add a bullet that asks for that check when no covering test is Cloud = true and the description does not mention a real-workspace run. Co-authored-by: Isaac --- .agents/skills/pr-review/SKILL.md | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/.agents/skills/pr-review/SKILL.md b/.agents/skills/pr-review/SKILL.md index fc00e7e5693..397ea485a14 100644 --- a/.agents/skills/pr-review/SKILL.md +++ b/.agents/skills/pr-review/SKILL.md @@ -139,6 +139,14 @@ on code the change adds or modifies, not on pre-existing issues next to it. should include the state after the action (list or print it again after a sync, a deploy or a migration) and the exact value the change is about, not just a successful exit. + - The fake server in `libs/testserver` only returns what its handlers were + taught, so a test that records the request, or a handler that echoes a + field back, shows what the CLI sends, not what the API accepts. When the + change sends a new field, a new value format (paths, identifiers, + durations) or a new call to a real API, and no test covering it has + `Cloud = true` (`Cloud` is inherited, so check `out.test.toml`) and the + description does not mention a run against a real workspace, ask whether + it was verified against one. - Are fixtures readable? Prefer generating archives and other binary inputs in the test script over committing opaque files. - Does the test reuse the repo's existing test helpers and scripts, and