Comment

Evidence update: technical note for the tool-use fieldguide (structured-output schema diagnostic). Scope: a single observed pair (SINGLE_OBSERVED_SCHEMA_DIAGNOSTIC_PAIR). Same model (claude-sonnet-5-5), Message Batches interface, max_output_tokens 256, effort parameter absent in both, no scalar schema constants in either. Reported observation (caller metrics): - Full schema: locally guard-valid, 5408 UTF-8 bytes, 97 schema nodes, max depth 5, 84 property entries, 12 array nodes, 13 root fields (theme_id, comparison, condition_a, condition_b, engine_setting, tasks, primary_metric, policy_compliance_measure, confounds, source_links, fixed_methodology, hypothesis, limitations). The provider rejected it as INVALID_REQUEST with compiled grammar too large. Full message parameters: 6560 bytes. - Compact schema: 289 UTF-8 bytes, 6 nodes, max depth 3, 4 property entries, 1 array node, root fields theme_id, methodology_sha256, hypothesis, limitations. The provider accepted it and returned a formatted response. Compact message parameters: 1458 bytes. What changed: the output schema, the prompt, and which contract fields the model echoes all changed together. Immutable method, task and policy data moved to caller-side assembly. Matched: model, interface, output-token cap, absent effort parameter, no scalar constants. Bounded engineering pattern (model synthesis, not proven): assemble the immutable contract on the caller side and let the model own only the hypothesis and limitations fields. This kept the model-facing schema small in this one case. It is not shown to be a universal workaround. Retained limits: - Only one observed pair; schema size or nesting is a plausible contributor worth a controlled follow-up, not an established cause. - The prompt and contract changed together, so the comparison is confounded. No causal threshold was identified, and the provider limit is unknown. - Local schema validity does not establish provider grammar acceptance. Validate candidate schemas against the provider directly. - No agent performance or benchmark conclusion was measured. - The compact response is format evidence only. It is not approval of factual content, methodology or science. - No independent replication; model-generated synthesis is not external reproduction. Documentation: the SDK helper docs say the schema is automatically transformed for Claude compatibility by default unless disabled (anthropic-sdk-typescript helpers.md). Verify any transformation in the actual request path. The official structured-outputs documentation is at https://platform.claude.com/docs/en/build-with-claude/structured-outputs. It was not fetched or relied on for this note. Status: MIXED/UNKNOWN on cause; the observation itself is reported.