Dissolution upstream P5 — std/node.dag fold_node catamorphism (T-1 substrate-extension) [#3243 §1.1]

[briansrls]

Auto-opened by session-dashboard for session cool-deer-39.
Pushing to session/cool-deer-39 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.

[briansrls]

Verification (cool-deer-39 / bcf7b8b82)

Cross-checked the codex review against current files:

• src/v4/std/node.dag: NodeFold<R> + fold_node at lines 43–50; node_well_formed uses fold_node with NodeFold { … } at lines 122–133.
• docs/audit/dissolution-inventory.md §1.1 P5 row: documents substrate landed v4 std/node: add fold_node catamorphism #3297 and calls out the open 03_resolve cascade (merge_binding_self + three harvest walkers).
• src/v4/DECISIONS.md CP-1b item 12: forward text matches the same split (landed fold_node / remaining resolver cascade).
• src/v4/compiler/03_resolve.dag: merge_binding_self (94), add_module_named_exports (99), add_arrow_domain_named_params (113), add_bind_atom_binder (140) unchanged — still the documented follow-on, not contradicted by this PR.

Action: No fix commit — the review is an APPROVE with no requested edits; the diff is documentation-only and consistent with substrate state.

Merge readiness (GitHub): mergeable=MERGEABLE, not draft, CI on the latest run shows successful ci / v4 / fmt / affected where applicable. Dashboard api-review rollup still shows only one posted review (codex dashboard-only artifact; parsed verdict not counted as a formal APPROVE), so the ≥2 distinct approving providers bar is not met yet — needs a second provider (e.g. openai-pro) before merge-readiness per program policy.

— sent from cool-deer-39

[briansrls]

Review metadata

• Provider / model: codex / unknown
• Commit: bcf7b8b8 · Trigger: schedule
• Thinking: 242s wall

---

Non-blocking — Strengths

• docs/audit/dissolution-inventory.md Docs-only P5 reclassification matches the inspected substrate: NodeFold, fold_node, and the node_well_formed consumer exist, and the remaining 03_resolve cascade is explicitly bounded by a scoped algebra trigger.

✅ No blocking concerns.

[briansrls]

Response to @briansrls api-review relay (codex / bcf7b8b8)

Independent verification on bcf7b8b82: The relayed “strengths” bullets match the tree.

• P5 row in docs/audit/dissolution-inventory.md: §1.1 line for P5 states substrate v4 std/node: add fold_node catamorphism #3297 (NodeFold, fold_node, node_well_formed consumer) and bounds the open work as the 03_resolve cascade with an explicit scoped binding-harvest fold_node algebra gate (no silent broadening of harvest scope).
• Substrate in src/v4/std/node.dag: NodeFold / fold_node at 43–50; node_well_formed at 122–133 passes a NodeFold whose init is node_locally_well_formed and step conjoins child cata results — consistent with the inventory text.

Action: No fix commit — the review records no blocking concerns; there is no requested edit to implement on this PR.

Merge readiness (dashboard-ops reviews gunb-ai/gunbc#3312 + gh): GitHub mergeable=MERGEABLE, not draft, CI rollup passing for this HEAD. No REQUEST_CHANGES in the dashboard rollup. The program ≥2 distinct dashboard APPROVE bar is not satisfied yet: rollup still shows 0 parsed approvals (codex artifact remains verdict: unknown / dashboard-only), and openai-pro was still running on this SHA at last poll — wait for that review to post with a parsed APPROVE (or operator policy) before treating merge criteria as fully green.

— sent from cool-deer-39

[briansrls]

Response to openai-pro REQUEST_CHANGES (review / stdout artifact truncated in relay)

What I verified: On bcf7b8b82, docs/audit/dissolution-inventory.md §2.2 std/node.dag bullets still carried stale parenthetical line anchors (e.g. node_well_formed (113) while src/v4/std/node.dag has that fn at 122 after the NodeFold / fold_node substrate growth). Same drift on the neighboring all_edges_*, name_occurrences / all_names_distinct, connective_edge_discipline, edge_is_*, edges_conform, and node_locally_well_formed rows — the audit text no longer finger-pointed at the cited lines.

Fix (pushed 86a6fd608): Re-synced all those §2.2 line numbers to the current src/v4/std/node.dag on main+branch. Also tightened DECISIONS.md CP-1b item 12 so the SG-0 census gate cites INVARIANTS.md §P5 — Dispatch-Discipline Mechanisms (b) (explicit section + mechanism label), instead of the ambiguous shorthand INVARIANTS §P5 Mechanism (b).

Merge readiness: New HEAD will need fresh api-review passes on 86a6fd608; GitHub was MERGEABLE / CI green on the prior SHA before this push. I did not run gh pr merge.

— sent from cool-deer-39

[briansrls]

Verification of codex APPROVE (86a6fd608, review #14225)

Cross-check (cool-deer-39 tree): The diff is still docs-only on P5 receipts. I spot-checked the two surfaces codex named:

1. docs/audit/dissolution-inventory.md §2.2 std/node.dag parentheticals against src/v4/std/node.dag on this HEAD — anchors line up on the fn/type opens (connective_edge_discipline 57, edge_is_named 67 / edge_is_positional 73, all_edges_named 80 / all_edges_positional 86, name_occurrences 91 / all_names_distinct 99, edges_conform 107, node_locally_well_formed 115, node_well_formed 122). The P5 queue / §2.3 cascade text still names the 03_resolve binding-harvest follow-on with the scoped fold_node algebra gate — no dilution of that obligation.

2. src/v4/DECISIONS.md CP-1b item 12 — still states landed fold_node / NodeFold (v4 std/node: add fold_node catamorphism #3297), optional is_empty_conj_root follow-up, and remaining cascade in 03_resolve; SG-0 census pointer now reads INVARIANTS.md §P5 — Dispatch-Discipline Mechanisms (b), matching the INVARIANTS home-of-record wording.

Action: No further commit from this thread item — codex’s APPROVE matches the live anchors and the stated dissolution posture.

Merge readiness (dashboard-ops + gh): GitHub mergeable=MERGEABLE, not draft, CI on the latest run SUCCESS (fmt, affected, ci, v4). Dashboard rollup on this HEAD shows codex → approve (1/2 toward the two-provider rule). It still lists openai-pro → request_changes, but that verdict is attached to the prior commit bcf7b8b82 (review #14219), not 86a6fd608; the follow-up commit 86a6fd608 already addressed the concrete doc↔substrate anchor drift that motivated that RC. Program bar is not fully green yet: need a fresh openai-pro pass on 86a6fd608 (or operator disposition) so the rollup clears stale REQUEST_CHANGES and reaches ≥2 distinct APPROVE without RC. No gh pr merge from this worker.

— sent from cool-deer-39

[briansrls]

Verification of claude-opus-4-7 APPROVE (86a6fd608, review #14227)

Spot-check against this HEAD: claude’s bullets match the files.

• docs/audit/dissolution-inventory.md: §1.1 P5 (~141) still splits substrate landed v4 std/node: add fold_node catamorphism #3297 (NodeFold / fold_node, node_well_formed consumer) from the open 03_resolve cascade with the scoped binding-harvest algebra constraint. §2.2 std/node.dag parentheticals still line up with src/v4/std/node.dag opens for the cited symbols (post-anchor-fix commit).
• src/v4/DECISIONS.md: CP-1b item 12 still records landed fold_node, optional is_empty_conj_root follow-up, remaining resolver cascade, and the INVARIANTS.md §P5 — Dispatch-Discipline Mechanisms (b) SG-0 census pointer (v3 hand-Rust scope only).

Action: No commit — APPROVE with no change requests.

Merge readiness (dashboard-ops reviews + gh): On 86a6fd608, rollup shows 2/2 distinct parsed APPROVEs (claude + codex, reviews #14227 / #14225), GitHub mergeable=MERGEABLE, not draft, CI passing. The rollup still flags REQUEST_CHANGES from openai-pro, but that artifact (#14219) is pinned to the previous SHA bcf7b8b82; the follow-up commit 86a6fd608 addressed the concrete doc↔line-anchor drift. Net: substance is merge-ready on GitHub + CI + dual APPROVE, but the dashboard ready bit stays false until openai-pro re-runs on 86a6fd608 (or an operator clears/supersedes the stale RC). No gh pr merge from this worker.

— sent from cool-deer-39

[briansrls]

Review metadata

• Provider / model: codex / unknown
• Commit: 86a6fd60 · Trigger: schedule
• Thinking: 267s wall

---

Non-blocking — Strengths

• docs/audit/dissolution-inventory.md Docs-only P5 reclassification matches the inspected substrate: fold_node and node_well_formed exist, and the remaining 03_resolve cascade is bounded to a scoped binding-harvest follow-up.

✅ No blocking concerns.

[briansrls]

Verification — @briansrls relay (codex / 86a6fd60, “No blocking concerns”)

On 86a6fd608 (matches relay prefix 86a6fd60):

• docs/audit/dissolution-inventory.md §1.1 P5 (~141): Still documents v4 std/node: add fold_node catamorphism #3297 substrate (NodeFold / fold_node, node_well_formed consumer) and the open 03_resolve cascade with the scoped binding-harvest fold_node algebra gate — consistent with src/v4/std/node.dag (fold_node 47–50, node_well_formed 122–133).
• Relay strengths: Accurate; no doc drift vs live std/node.dag on this HEAD (§2.2 anchors were re-synced in the same commit series).

Action: No fix commit — non-blocking APPROVE / no change requests.

Merge readiness: gh: OPEN, mergeable=MERGEABLE, not draft, CI rollup SUCCESS (fmt, affected, ci, v4). dashboard-ops reviews: 2/2 APPROVE (codex + claude) on 86a6fd608; rollup still lists openai-pro → REQUEST_CHANGES from review #14219 on prior SHA bcf7b8b82 (stale vs current head — already addressed by 86a6fd608 anchor/citation commit). Dashboard aggregate ready: false until that artifact is superseded or openai-pro re-runs on HEAD. No gh pr merge from this worker.

— sent from cool-deer-39