Improve skills based on CAO benchmark run 2 findings#89
Merged
Conversation
Address 4 friction clusters from the May 2026 benchmark: - OAuth auth: add PAT limitation troubleshooting and note on apps logs - Model Serving: add endpoint readiness section with state transitions - Off-platform: strengthen anti-pattern warning (not Databricks Apps) - Lakehouse Sync: make UI-only limitation more prominent Co-authored-by: Isaac
…times Co-authored-by: Isaac
pkosiec
force-pushed
the
pkosiec/benchmark-improvements
branch
from
7454094 to
861875f
Compare
pkosiec
requested review from
a team,
lennartkats-db and
simonfaltum
as code owners
Contributor
|
Latest dev
(Prod and dogfood nightlies have unrelated infra issues that pre-date this PR — pointed those out separately.) |
keugenek
approved these changes
May 26, 2026
Co-authored-by: Isaac
lennartkats-db
approved these changes
May 26, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Addresses 4 friction clusters identified in the CAO (Coding Agent Optimization) benchmark run 2 (May 21-22, 2026; Opus 4.7 + Codex 5.5, 25 tasks).
Changes
P2: OAuth auth guidance (Cluster A -- 16/25 tasks affected)
databricks-core/SKILL.md: Add troubleshooting row for"OAuth Token not supported for current auth type pat"directing todatabricks auth logindatabricks-apps/SKILL.md: Note thatdatabricks apps logsrequires OAuth in Post-Deploy VerificationP4: Model Serving endpoint readiness (Cluster F -- regression on both models)
databricks-model-serving/SKILL.md: Add "Endpoint Readiness" subsection with state transitions, provisioning times, and poll-before-query guidance. Enhance troubleshooting forNOT_READYstate.P5: Off-platform anti-pattern (Drizzle task -- Codex scores 35)
databricks-lakebase/references/off-platform.md: Add callout that off-platform apps are NOT Databricks Apps. Add "Running Locally" section with standard Node.js commands.databricks-lakebase/SKILL.md: Strengthen off-platform reference description.P6: Lakehouse Sync UI-only limitation (Cluster G)
databricks-lakebase/references/lakehouse-sync.md: Strengthen warning: "Do NOT attempt to automate this step."databricks-lakebase/SKILL.md: Add "(UI-only)" annotation to Lakehouse Sync reference entry.Context
The benchmark also identified these clusters that are addressed elsewhere:
package-lock.jsonentry)experimental aitools installdeprecation): Addressed in PR docs: track aitools rename to top-level command #60Test plan
python3 scripts/skills.py validatepassesThis pull request and its description were written by Isaac.