@@ -148,17 +148,14 @@ jobs:
148148 TRIGGER_COMMENT_ID : ${{ github.event.comment.id }}
149149 MODEL_ENDPOINT : https://${{ secrets.DATABRICKS_HOST }}/serving-endpoints/databricks-claude-opus-4-8/invocations
150150 DATABRICKS_TOKEN : ${{ secrets.DATABRICKS_TOKEN }}
151- # Live-warehouse connection env, provisioned for parity with the author
152- # run + code-coverage.yml (job is in `environment: azure-prod`). NOTE:
153- # unlike the author phase, the followup prompt
154- # (.bot/prompts/engineer-followup/system.md) tells the agent to run only
155- # the mocked `tests/unit` suite and NOT tests/e2e. These vars are wired
156- # for consistency with the author run, not an E2E step the followup runs;
157- # if followups should ever repair/run E2E repros, update that prompt too.
158- DATABRICKS_SERVER_HOSTNAME : ${{ secrets.DATABRICKS_HOST }}
159- DATABRICKS_HTTP_PATH : ${{ secrets.TEST_PECO_WAREHOUSE_HTTP_PATH }}
160- DATABRICKS_CATALOG : peco
161- DATABRICKS_USER : ${{ secrets.TEST_PECO_SP_ID }}
151+ # NOTE: no live-warehouse connection env (DATABRICKS_SERVER_HOSTNAME /
152+ # DATABRICKS_HTTP_PATH / DATABRICKS_CATALOG / DATABRICKS_USER) is wired
153+ # here. The followup prompt (.bot/prompts/engineer-followup/system.md)
154+ # runs only the mocked `tests/unit` suite and forbids tests/e2e, so the
155+ # agent never consumes those vars — provisioning them would add live
156+ # credentials to an LLM-driven step for no functional benefit. If
157+ # followups should ever repair/run E2E repros, add them back here AND
158+ # update that prompt.
162159 RUNNER_TEMP : ${{ runner.temp }}
163160 # The agent's working tree AND the .bot/ lookup root. run.py resolves
164161 # the config at <REPO_ROOT>/.bot/config.yaml. The engine has NO path
0 commit comments