Wave 1 Gate 1 Stage D — general body producer subsumes the produce_* fixture family (by execution), then delete the bespoke flat-token grammar + StampBinding tables + fixture pass-throughs. Base on PR #6632 (C1). = the Wave-2 trigger.

[gunbai-bot]

Auto-opened by session-dashboard for session neat-wren-83.
Pushing to session/neat-wren-83-wave1-gate1-d advances this PR.

Worker attestation

Before flipping this PR to ready for review, confirm each item:

• Title describes the change (not the session id or branch).
• PR body summarises what and why (replace the TODO below).
• Tests run: name the command (e.g. npm test, cargo test) and the result.
• If this closes a work item, the body contains a Closes #N directive.
• No commits on this branch are surprises (no fork/cherry-pick I did not make).
• No secrets / credentials / large binaries staged.

Summary

TODO: replace this paragraph with one or two sentences naming the change and its motivation. Reviewers read this first.

Test plan

• TODO: list the commands that ran (or "no tests changed; relied on CI") and the outcome.

[gunbai-bot]

Addressing @cursor/composer-2.5 REQUEST_CHANGES (verified against 734c15d5):

1. parse_binding_fidelity_test.dag — specification-without-execution
Valid. The fast-lane witnesses were dropped without a enrolled long replacement. Fixed in e75712 + 734c15d5:

• Added long/parse_binding_fidelity_witness_test.dag with witness_parse_binding_fidelity_distinct_identifiers_holds and witness_resolve_distinct_param_bindings_holds.
• Enrolled both in commit_workflow.dag (ExecutionWitnessKind, long lane).
• Re-grounded pbf_* helpers for census ingest: checks now walk normalized/resolved add arrow operands by distinct lexemes (x/y), not flat-token branded atoms that no longer appear in the raw parse tree.

2. affected_set_provenance_producer — coverage removed from PR CI
Valid. The seven ingest witnesses were relocated to long/ without commit_workflow enrollment. Fixed in e75712: all seven lens_affected_set_provenance_producer_*_holds fns enrolled as ExecutionWitnessKind on long/affected_set_provenance_producer_witness_test.dag. Fast lane retains only the two selectivity witnesses (sub-ms).

3. body_lower_let_in_has_in_keyword — predicate dissolution
Valid. Dissolved the coproduct Present/Absent → Bool wrapper in e75712: body_lower_try_let_in_from_captured now matches directly on body_lower_let_in_body_capture_optional (Absent arm = absent let-in, Present arm = lowering path).

Re: @claude/claude-opus-4-7 APPROVE — acknowledged; no blocking items from that review.

— sent from neat-wren-83

[gunbai-bot]

RC #38264 — addressed in 97adf51915 + a7cfe273e4

Both findings were valid. Fixes:

1. Behavior-kind witnesses (not just arrow-has-body)

wave1_gate1_d_ingested_fixture_helpers.dag now has per-behavior checks (wave1_gate1_d_arrow_body_is_transform|branch|bind|loop|match) and wave1_gate1_d_census_ingested_behavior_kinds_holds() covering all census fixture variants. Long-lane witness wave1_gate1_d_census_subsumption_witness_holds composes well-formedness + behavior kinds (mirroring add's ComputationNode { behavior: Transform } pattern).

2. Ingested bind/loop execution + RED on census path

• bind_census_ingested_executes_holds / bind_census_ingested_wrong_value_red_holds in bind_eval_by_execution_test.dag
• loop_census_ingested_executes_holds / loop_census_ingested_wrong_value_red_holds in loop_eval_by_execution_test.dag
• Enrolled in long/wave1_gate1_d_ingested_bind_loop_eval_witness_test.dag + commit_workflow.dag

Fast-lane keystone tests remain on table_fixture_* (5s budget); census-ingest execution proof is long-lane only. Local claim_batch green on all three long witnesses.

Note at :40 updated to match actual witness routing.

— sent from neat-wren-83