coresearch: CE64 + Moore + masking wiring map, six probes pinned

[AdaWorldAPI]

What

This PR records the coresearch council's answer to stockfish-rs inventory items O1–O5, which ask how to wire thinking onto CausalEdge64 and the mask algebra. It also adds six runnable probes, in the operator's priority order: CE64, Moore, masking, with the shader driver kept in the loop.

Board entry: .claude/board/entries/2026-10-08-coresearch-ce64-moore-masking-wiring.md. It contains the exploration map (PROBE / PARK / SKIP) and states what was not searched. The council ratifies nothing.

The council's main finding: every idea that survives goes beside CE64, not inside it. The three carriers (CE64 per edge, ThoughtCtx per thought, MailboxSoA.meta and masks per population) already exist; what is missing is the wiring between them.

Probes

Each probe is test-pinned two-sided, and each pinned behaviour was disable-verified red.

probe	where	result
CF-MANTISSA	causal-edge/tests/ce64_op_contract.rs	A weight carrying Counterfactual (−6) runs the Synthesis average (128, 128), then gets re-stamped −6.
W-PRESERVE	same file	forward returns W = 0 and bits 59..63 = 0 even when both operands carry them. learn and the single-field setters preserve both.
PREFILTER-MASK	cognitive-shader-driver/examples/meta_prefilter_mask_probe.rs (test = true)	MetaFilter lowers to a row mask built only from ternary_match_u32_to_mask_under, with no field extraction. 400 filters × 4099 rows give the same rows as the shipped selection-vector meta_prefilter.
MOORE-PLANES	moore_plasticity_probe.rs (+2 tests)	The Moore schedule equals grid & shift(grid, −d) via mask_shift_morton. The palette fold stays a value-plane LUT fold.
CHAIN-CONF / STAMP-GATE	lance-graph-planner/tests/chain_confidence.rs	On a 5-hop deduction chain, replay_step raises confidence 200 → 224 → 237 and then saturates, while forward's own deduction falls. Self-revision raises confidence. A weight with zero confidence still adds confidence (128 → 137, from the 16-bin revise table).

No production code changes. The probes pin current behaviour, so a fix has to flip a test on purpose.

Checks

• cargo test -p causal-edge passes, including --no-default-features (the new file is gated to v2).
• cargo test -p cognitive-shader-driver --example meta_prefilter_mask_probe and --example moore_plasticity_probe pass.
• cargo test -p lance-graph-planner --test chain_confidence passes.
• cargo clippy -p causal-edge --all-targets -D warnings reports 9 errors, all pre-existing in src/. The new test file is clean.
• entries_index.py --write and supersession_index.py regenerated.

🤖 Generated with Claude Code

https://claude.ai/code/session_01EFw2WdKr1oxvaKCJC2ua2R

---

Generated by Claude Code

Summary by CodeRabbit

• Tests
  • Added coverage for causal-edge operation consistency, confidence behavior, and preservation of key values.
  • Added checks for filtering results and directional neighbor calculations across varied inputs.
• Documentation
  • Expanded the research index with findings on processing, masking, and state handling.

[coderabbitai]

📝 Walkthrough

Walkthrough

The change adds tests for CE64 operations, confidence replay, metadata prefilter masks, and Moore direction masks. It also adds a board entry with probe findings and updates the board indexes.

Changes

Probe Validation and Findings

Layer / File(s)	Summary
CE64 operation contracts crates/causal-edge/tests/ce64_op_contract.rs	V2-layout-gated tests check forward dispatch and repeatability, field preservation by setters and learn, and fields returned by forward and syllogize.
Chain confidence and replay crates/lance-graph-planner/tests/chain_confidence.rs	Tests compare confidence traces for forward deduction and replay. Additional tests check replay with identical edges and zero-confidence weights.
Metadata prefilter mask probe crates/cognitive-shader-driver/Cargo.toml, crates/cognitive-shader-driver/examples/meta_prefilter_mask_probe.rs	The registered probe lowers filter clauses to row masks and compares its selected rows with BindSpace::meta_prefilter. Tests cover randomized inputs, threshold boundaries, and window selection.
Moore direction-mask checks crates/cognitive-shader-driver/examples/moore_plasticity_probe.rs	Tests compare Morton-shifted direction masks with the closed-form schedule and check population counts and swapped-axis behavior.
Probe findings and indexes .claude/board/entries/2026-10-08-coresearch-ce64-moore-masking-wiring.md, .claude/board/entries/README.md, .claude/board/SUPERSESSION-INDEX.md	The board entry records findings, planned probes, and unsearched topics. The entry count and BindSpace crate-file count are updated.

Priority: ⬇️ Low

Estimated code review effort: 3 (Moderate) | ~20 minutes

Change: Other

Suggested reviewers: claude

Merge Risk: ⚪ Minimal · up to b454d

The reported full-size measurement and the automated equivalence tests serve different purposes. No issue established here needs to block merging.

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 63.41% which is insufficient. The required threshold is 80.00%. Docstring coverage is scoped to functions touched by this diff. Analyzed 41 functions across 4 files. (4 skipped: …	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title clearly identifies the coresearch wiring map and the six pinned probes, which matches the main changes in the pull request.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

Full details: Docstring Coverage

Explanation

Docstring coverage is 63.41% which is insufficient. The required threshold is 80.00%. Docstring coverage is scoped to functions touched by this diff. Analyzed 41 functions across 4 files. (4 skipped: 4 unsupported.)

• Fix all pre-merge checks with AI

✨ Finishing Touches 💡 1

📝 Generate docstrings 💡

• Commit to this branch
• Create a new PR

• Autopilot · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

Warning

Billing warning: we have not been able to collect payment for this subscription for more than 72 hours. Please update the payment method or pay any pending invoices in Billing to avoid service interruption.

---

_{A rabbit checks each mask with care
Then hops through fields from here to there
The Moore lanes line up in rows
Replay confidence climbs and grows
CE64 tests record what shows}

_{Comment @coderabbitai help to get the list of available commands.}

[chatgpt-codex-connector]

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-10-08T06:18:02.589411Z	98cc9d5	Draft marked ready

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 98cc9d5bb6

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[coderabbitai]

🔇 Additional comments (5)

crates/causal-edge/tests/ce64_op_contract.rs (1)

1-175: LGTM!

.claude/board/entries/2026-10-08-coresearch-ce64-moore-masking-wiring.md (1)

1-162: LGTM!

.claude/board/entries/README.md (1)

28-28: LGTM!

Also applies to: 36-36

.claude/board/SUPERSESSION-INDEX.md (1)

17-17: 📐 Maintainability & Code Quality

The count updates are expected for a generated index. CLAUDE.md states that SUPERSESSION-INDEX.md is generated, must not be hand-edited, and must be regenerated when its inputs change. The append-only rule does not prohibit regenerated measurement changes.

crates/cognitive-shader-driver/examples/meta_prefilter_mask_probe.rs-181-182 (1)

181-182: 🎯 Functional Correctness | 🟡 Minor | ⚡ Quick win

⚠️ Unverified finding
Verification ran but could not confirm this finding. It is shown for review, not as a verified issue.

Run the full-size probe in a test.

The reported cargo test --example meta_prefilter_mask_probe check does not call main(): Cargo replaces it with the test harness. The test compares 200 filters over 1031 rows, so that check does not verify the reported 400-filter, 4099-row result. Add a #[test] that calls main(), or move the full-size comparison into a test. Keep test = true so the existing tests still run. (doc.rust-lang.org)

---

ℹ️ Review info

⚙️ Run configuration

• Configuration used: Organization UI
• Review profile: CHILL
• Plan: Essentials
• Run ID: 89232ade-1490-47aa-8f81-a41577cfc8d8

📥 Commits

Reviewing files that changed from the base of the PR and between bcc9f50 and b454d51.

📒 Files selected for processing (8)

• .claude/board/SUPERSESSION-INDEX.md
• .claude/board/entries/2026-10-08-coresearch-ce64-moore-masking-wiring.md
• .claude/board/entries/README.md
• crates/causal-edge/tests/ce64_op_contract.rs
• crates/cognitive-shader-driver/Cargo.toml
• crates/cognitive-shader-driver/examples/meta_prefilter_mask_probe.rs
• crates/cognitive-shader-driver/examples/moore_plasticity_probe.rs
• crates/lance-graph-planner/tests/chain_confidence.rs

Included review availability: This review used your included allowance. 0 included reviews remain after this review. Your included PR review attempts over the past 7 days set your current allowance at 1 review per hour.