fix(learning): repair MODEL_ENDPOINT 400 + bump engine to 654a31d0 - #905
Merged
Conversation
Move all bots to engine main (654a31d0), 30 commits ahead of the current pin. This repo already runs the modern daily-cron retrospective + closed loop; this brings the pinned engine in line with the sibling driver repos (kernel + the go/nodejs/jdbc/odbc learning PRs), whose engine was bumped to the same SHA. The main addition since d05dcb11 is per-bot model selection (default opus-4.8). SDK/CLI locked versions (0.2.102 / 2.1.61) are unchanged at 654a31d0, so this is a SHA-only bump — no install-bot-engine edit needed. Co-authored-by: Isaac Signed-off-by: eric-wang-1990 <e.wang@databricks.com>
eric-wang-1990
temporarily deployed
to
azure-prod
August 13, 2026 05:13 — with
GitHub Actions
Inactive
eric-wang-1990
temporarily deployed
to
azure-prod
August 13, 2026 05:13 — with
GitHub Actions
Inactive
There was a problem hiding this comment.
Pull request overview
Updates the pinned bot engine commit used by the shared .github/actions/bot-prelude composite action, aligning this repo’s bot workflows with the newer engine SHA referenced by sibling repositories.
Changes:
- Bump the
engine-refinput default fromd05dcb11…to654a31d0…in the bot prelude composite action.
💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What
Two changes to the engineer-bot learning flow:
Fix the daily-failing cron (
MODEL_ENDPOINT). The learning workflow used.../serving-endpoints/anthropic/invocations, which is not translated correctly:sdk_agent.translate_endpointearly-returns on URLs already containing/serving-endpoints/anthropic, keeping the trailing/invocations. The CLI then appends/v1/messages→.../serving-endpoints/anthropic/invocations/v1/messages→400 Unsupported native API path. This repo's learning cron has failed every scheduled run with exactly this error (e.g. run 31624062797). Switch to the concrete.../serving-endpoints/databricks-claude-opus-4-8/invocationsform thatreviewer-bot.yml/engineer-bot.ymluse successfully.Bump the engine pin
d05dcb11→654a31d0(enginemain, 30 commits): brings per-bot model selection. SDK/CLI (0.2.102/2.1.61) unchanged at that SHA — SHA-only.Why together
The endpoint bug is why the retrospective never actually ran here despite the flow being wired up. Fixing it + bumping to the current engine gets the daily learning extraction working on
mainfor the first time. The same endpoint fix is going out to the sibling driver repos' learning PRs.Note
The effective model is engine-owned (
repo_conventions.engineer_bot_model()=databricks-claude-opus-4-8[1m], override viaENGINEER_BOT_MODEL_OVERRIDE); the model segment in the URL is discarded bytranslate_endpoint, so this is purely a routing/base-URL fix.This pull request and its description were written by Isaac.