P3: wire the rung ascent loop + dedup RungLevel (D-TRI-6)#708
Merged
Conversation
Executes the §3a Maslow-pyramid wiring that E-MASLOW-PYRAMID-OF-COGNITION-1 named as the missing piece. Four deliverables, all verified green. (D) Rung ascent loop — the driver already held a per-driver RungElevator and called on_gate() once per dispatch, but the rung never reached projection selection (stage [3] cascade used req.layer_mask verbatim). Now the elevator's current level (advanced by the previous cycle's gate) selects this cycle's cascade breadth via rung_widened_layer_mask, closing the loop across cycles. CORRECTNESS GATE resolved: p64 cascade's layer_mask is an 8-bit PREDICATE-PLANE mask (CAUSES..BECOMES), NOT the 3-bit SPO projection mask causal_mask_bits() returns — so the rung's Pearl level selects a predicate-plane widen-set applied as a UNION gated on elevation-above-base: identity at base (zero regression), superset-monotone above. D-TRI-6 settlement probe green (BLOCK ascends, FLOW relaxes to base, mask widens then returns to identity). (A) Dedup — thinking-engine's variant-identical RungLevel collapsed to a re-export of the contract type (added as_u8() to the contract; a discriminant-parity test is the measure step). Removed dead should_elevate, LayerId (zero consumers), and Agent.current_rung (write-once Surface, never mutated — the loop lives on the driver's elevator). (B) scenario.rs Pearl-numbering doc fix (interventions = rung 2, not 3). (C) nars_engine deprecation silenced with justified #[allow(deprecated)] (v2 inference_type() already routes through from_mantissa; temporal() inert) — planner clippy -D warnings exit 0. Verified: contract 905 tests + parity green + clippy exit 0; planner 215 + clippy exit 0; driver 103 + both D-TRI-6 probes green; thinking-engine lib builds. System-1 apply_delta arm left test-only (no emergence/coherence scalar in the driver hot path; fabricating one is forbidden). Board: STATUS_BOARD D-TRI-6 (ascent loop wired), EPIPHANIES E-RUNG-ASCENT-WIRED-1, TECH_DEBT TD-THINKING-ENGINE-EXCLUDED-DEBT-1 + TD-ONTOLOGY-CLIPPY-DEBT-1 (pre-existing debts surfaced, out of P3 scope). Co-Authored-By: Claude <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Awg6TXocHcwTtc6eGsHcdD
Bugbot couldn't run - usage limit reachedBugbot is counted against Cursor usage for this user or team, and this run hit a usage or spend limit. A user or team admin can review and increase usage limits in the Cursor dashboard. (requestId: serverGenReqId_88b8ec0e-6c79-45eb-94d8-50e1389acca4) |
|
Warning Review limit reached
Next review available in: 44 minutes Enable usage-based reviews in Billing to review now. Otherwise, wait until the next included review is available. How can I continue?After more reviews become available, a review can be triggered using the To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based reviews. How do review limits work?CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability. For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window. Please refer docs for additional details. Review details⚙️ Run configurationConfiguration used: Organization UI Review profile: CHILL Plan: Pro Plus Run ID: 📒 Files selected for processing (2)
📝 WalkthroughWalkthroughThe PR wires persistent ChangesRung ascent contracts and integration
Estimated code review effort: 4 (Complex) | ~45 minutes Sequence Diagram(s)sequenceDiagram
participant ShaderDriver
participant RungElevator
participant CognitiveShader
ShaderDriver->>RungElevator: Align to requested base rung
RungElevator-->>ShaderDriver: Return current rung level
ShaderDriver->>CognitiveShader: Cascade with effective layer mask
ShaderDriver->>RungElevator: Advance on gate result
Possibly related PRs
Suggested reviewers: Poem
🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
AdaWorldAPI
marked this pull request as ready for review
July 17, 2026 12:06
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 184efcc153
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".
Comment on lines
+604
to
+606
| // base; sustained-FLOW relaxes back to it). The base-reset already ran | ||
| // at the top of `run()` where the level was read for the cascade, so | ||
| // `elevator.base == req.rung` holds here. The resulting level is what the |
There was a problem hiding this comment.
When the same ShaderDriver handles concurrent dispatches with different ShaderDispatch::rung values, the base reset done at the top of run() is not protected until this later update. Another dispatch(&self) can acquire the rung_elevator write lock in between and reset base, so the assumption that elevator.base == req.rung no longer holds; this dispatch’s gate is then counted against the other request’s base and can make the next cascade use the wrong widened mask. Keeping the reset in the same critical section as on_gate avoids cross-request rung leakage.
Useful? React with 👍 / 👎.
Owner
Author
There was a problem hiding this comment.
Correct — fixed in c793556. The per-driver RungElevator is a &self RwLock singleton, so the split I introduced (read level at the top of run(), on_gate at the bottom, in two separate critical sections) opened exactly the window you describe: a concurrent dispatch with a different req.rung could reset base in between, and this cycle's gate would then count against the wrong base. Restored the if elevator.base != req.rung { reset } inside the same critical section as on_gate (as the pre-PR code had it), keeping reset+advance atomic. Parity with the original atomicity; a genuinely per-request elevator is the triangle-arc mailbox-SoA design, out of scope here. Driver rebuilt: 103 tests pass, both D-TRI-6 probes green.
Generated by Claude Code
Codex correctly flagged a concurrency window I introduced. The per-driver RungElevator is a &self singleton behind a RwLock, so concurrent dispatches share it. My change read the elevator's level at the top of run() (for the cascade mask) and advanced it via on_gate at the bottom, in TWO separate critical sections — dropping the base-reset from the bottom block on the assumption elevator.base == req.rung still held. Under concurrency a sibling dispatch with a different req.rung can reset base in the window between, so on_gate then counts this cycle's gate against the wrong base and the next cascade widens with the wrong mask. Fix: restore the reset-if-mismatch inside the on_gate critical section (as the original code had it), keeping reset+advance atomic. Parity with the pre-PR atomicity; the deeper per-request elevator is the triangle-arc mailbox-SoA design, not this PR. Driver rebuilt: 103 tests pass, both D-TRI-6 probes green. Co-Authored-By: Claude <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Awg6TXocHcwTtc6eGsHcdD
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 6
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In @.claude/board/EPIPHANIES.md:
- Around line 1-11: Move the new dated section in EPIPHANIES.md from the
beginning to the end of the existing entries, preserving all prior content and
ordering unchanged. Treat the file as an append-only audit ledger and append
this update rather than rewriting or prepending it.
In @.claude/board/STATUS_BOARD.md:
- Line 12: Preserve the original D-TRI-6 row in .claude/board/STATUS_BOARD.md
and append a dated status update instead of editing it in place. Move the new
debt entries in .claude/board/TECH_DEBT.md from lines 3-10 to the ledger’s
append point. Treat all .claude/board/*.md files as append-only audit ledgers.
In @.claude/board/TECH_DEBT.md:
- Around line 3-10: Move the new TD-THINKING-ENGINE-EXCLUDED-DEBT-1 and
TD-ONTOLOGY-CLIPPY-DEBT-1 records from the top of TECH_DEBT.md to the file’s
append point after the existing ledger history. Preserve their content and
ordering, ensuring .claude/board/*.md remains append-only.
In `@crates/cognitive-shader-driver/src/driver.rs`:
- Around line 240-249: Update the dispatch-state handling in run() around
rung_elevator so reading the current rung, computing the widened layer mask,
applying the gate, and updating the elevator occur within one serialized
transaction. Prevent concurrent dispatches from resetting or advancing another
request’s base, while releasing the lock before any sink callbacks to avoid
holding it across reentrant callback execution.
- Around line 602-614: Move the `rung_elevator` write-lock and
`elevator.on_gate(gate)` transition from the current post-sink block to
immediately after `gate` is decided, before either sink callback or its early
returns. Ensure both `BLOCK` and `FLOW` outcomes advance the persistent elevator
even when sink processing terminates the dispatch, while preserving the
resulting rung used by subsequent dispatch logic.
In `@crates/lance-graph-planner/src/cache/nars_engine.rs`:
- Around line 503-511: The v2 SpoHead round-trip must not copy reclaimed bits
from CausalEdge64::temporal() into SpoHead.temporal. Update the round-trip
conversion around the #[allow(deprecated)] site to preserve temporal only for
the v1 layout, or use the documented v2 sentinel, while retaining current v1
behavior; add a field-isolation test covering the v2 path.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro Plus
Run ID: b23ff902-36f8-4971-afab-ba7625a0b391
📒 Files selected for processing (10)
.claude/board/EPIPHANIES.md.claude/board/STATUS_BOARD.md.claude/board/TECH_DEBT.mdcrates/cognitive-shader-driver/src/driver.rscrates/lance-graph-contract/src/cognitive_shader.rscrates/lance-graph-contract/src/scenario.rscrates/lance-graph-planner/src/cache/nars_engine.rscrates/thinking-engine/src/cognitive_stack.rscrates/thinking-engine/src/persona.rscrates/thinking-engine/src/world_model.rs
…2 temporal sentinel Two valid review findings fixed: 1. (CodeRabbit, driver.rs) The sink early-returns (on_resonance/on_bus returning false, lines 556/579) bypassed the on_gate block at the bottom of run(), so a sink that stops processing would freeze the rung ascent ladder — now that the elevator drives the cascade this is observable. Moved the on_gate transition to immediately after gate is decided, BEFORE the sink callbacks, so every dispatch that reaches a gate advances the elevator. Verified no early return exists between gate-decision (l.454) and the sinks. The base-reset stays inside the same critical section (the codex atomicity fix, c793556). 2. (CodeRabbit, nars_engine.rs) from_causal_edge copied edge.temporal() into SpoHead.temporal. causal-edge defaults to causal-edge-v2-layout, where bits 52-63 are reclaimed (W-slot/lens/spare) and temporal() returns GARBAGE (edge.rs:551 doc). Per I-LEGACY-API-FEATURE-GATED, temporal is now the documented 0 sentinel (temporal is structural: SpoWitnessChain / AriGraph). inference_type() keeps #[allow(deprecated)] — it returns the CORRECT value under v2 (routes through from_mantissa), deprecated only as an API nudge. Added field-isolation test from_causal_edge_temporal_is_v2_sentinel_not_reclaimed_bits (raw reclaim-zone bits -> temporal() garbage -> SpoHead.temporal == 0). Verified: driver 103 tests + both D-TRI-6 probes green; planner 215 + the new isolation test green; planner clippy --all-targets exit 0. Co-Authored-By: Claude <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Awg6TXocHcwTtc6eGsHcdD
AdaWorldAPI
pushed a commit
that referenced
this pull request
Jul 17, 2026
Evaluation + wave plan for feeding a new query-side crates/graphrag and OGAR's ogar-doc from the existing SoA + CausalEdge64 + (queued) EpisodicWitness64 substrate, aligned to #708 / D-TRI-6 (the rung-ascent loop) and the v3 waves. Key rulings the plan encodes: - Direction reframe (assembler-vs-storage boundary): OGAR owns doc ingestion + the document 0x080B mint (ogar-from-docv1 -> ogar-doc-ir -> ogar-doc, built); lance-graph provides contract types OGAR consumes + query. graphrag is query-side ONLY, reads the calcified substrate, never ingests. A lance-graph doc-transcode crate is REJECTED (duplicates ogar-from-docv1, inverts the dep). - REUSE the substrate, BUILD only 3 gaps: hierarchical Leiden, HippoRAG-PPR reset-distribution, BM25. Extraction/SPO/fact-store/vector-CAM-PQ already exist; causal_edge::CausalEdge64 is the SPO edge (not the thinking-engine variant); EpisodicWitness64 is a queued column (today WitnessTable + EpisodicEdges64). graphrag-rs = REUSE-AS-REFERENCE (E-V3-GRAPHRAG-INV-1). - Retrieval = rung ascent (#708): dispatches through RungElevator + canonical RungLevel; BLOCK ascends and widens the CAUSES..BECOMES predicate-plane mask, FLOW relaxes to base. The graph is load-bearing because surprise ascends the elevator. Maps onto the triangle-tenants Maslow pyramid (D-TRI). - Probe-first: G0 P-GRAPH-LOADBEARING (graph vs vector-only) gates all Leiden/PPR code. Composes atop oxigraph-arigraph-cognitive-shader-soa-merge-v1. No bytes land; D-GR-5 (OGAR seam) mint-gated + baton-audited. Board hygiene: INTEGRATION_PLANS prepend in the same commit. Co-Authored-By: Claude <noreply@anthropic.com>
AdaWorldAPI
pushed a commit
that referenced
this pull request
Jul 17, 2026
Evaluation + wave plan for feeding a new query-side crates/graphrag and OGAR's ogar-doc from the existing SoA + CausalEdge64 + (queued) EpisodicWitness64 substrate, aligned to #708 / D-TRI-6 (the rung-ascent loop) and the v3 waves. Key rulings the plan encodes: - Direction reframe (assembler-vs-storage boundary): OGAR owns doc ingestion + the document 0x080B mint (ogar-from-docv1 -> ogar-doc-ir -> ogar-doc, built); lance-graph provides contract types OGAR consumes + query. graphrag is query-side ONLY, reads the calcified substrate, never ingests. A lance-graph doc-transcode crate is REJECTED (duplicates ogar-from-docv1, inverts the dep). - REUSE the substrate, BUILD only 3 gaps: hierarchical Leiden, HippoRAG-PPR reset-distribution, BM25. Extraction/SPO/fact-store/vector-CAM-PQ already exist; causal_edge::CausalEdge64 is the SPO edge (not the thinking-engine variant); EpisodicWitness64 is a queued column (today WitnessTable + EpisodicEdges64). graphrag-rs = REUSE-AS-REFERENCE (E-V3-GRAPHRAG-INV-1). - Retrieval = rung ascent (#708): dispatches through RungElevator + canonical RungLevel; BLOCK ascends and widens the CAUSES..BECOMES predicate-plane mask, FLOW relaxes to base. The graph is load-bearing because surprise ascends the elevator. Maps onto the triangle-tenants Maslow pyramid (D-TRI). - Probe-first: G0 P-GRAPH-LOADBEARING (graph vs vector-only) gates all Leiden/PPR code. Composes atop oxigraph-arigraph-cognitive-shader-soa-merge-v1. No bytes land; D-GR-5 (OGAR seam) mint-gated + baton-audited. Board hygiene: INTEGRATION_PLANS prepend in the same commit. Co-Authored-By: Claude <noreply@anthropic.com>
AdaWorldAPI
pushed a commit
that referenced
this pull request
Jul 17, 2026
…ate (§10) Verified every reuse/build assumption against the tree at 8d3209c (#708 merged). Verdict: FEASIBLE, two small non-blocking caveats, no circular deps. Confirmed reachable (pub): RungElevator (pub struct/fields + new + on_gate) and RungLevel/GateDecision/MailboxSoaView + CausalEdge64 all from ZERO-DEP crates; TripletGraph exposes the exact walk (get_associated multi-hop, intervene_on Pearl-2 apex, detect_contradictions = the BLOCK trigger, revise_with_evidence), spo_bridge::promote_to_spo pub, hdr_pagerank + ScentCsr::spmv pub, cam_pq pub. The rung loop maps 1:1 onto existing methods. Caveat 1 (drives a feature gate): the graph legs live in lance-graph core = drags datafusion/lance/arrow (19 heavy-dep lines); the rung-ascent core is zero-dep. => default graphrag = LIGHT (rung ascent + BM25 + in-memory PPR + G0 probe); feature graphrag-lance = HEAVY graph legs. §4 crate topology updated. Caveat 2: rung_widened_layer_mask is private (driver.rs:701) => make pub / move to contract, or replicate the pure fn (secondary leg; ascent works without it). Genuine BUILD confirmed: hierarchical Leiden (jc louvain is example-only), HippoRAG-PPR reset-distribution atop hdr_pagerank, BM25. Next step needs no upstream change: scaffold the light crate + cargo-check compile-proof, then G0. Co-Authored-By: Claude <noreply@anthropic.com>
AdaWorldAPI
pushed a commit
that referenced
this pull request
Jul 17, 2026
Evaluation + wave plan for feeding a new query-side crates/graphrag and OGAR's ogar-doc from the existing SoA + CausalEdge64 + (queued) EpisodicWitness64 substrate, aligned to #708 / D-TRI-6 (the rung-ascent loop) and the v3 waves. Key rulings the plan encodes: - Direction reframe (assembler-vs-storage boundary): OGAR owns doc ingestion + the document 0x080B mint (ogar-from-docv1 -> ogar-doc-ir -> ogar-doc, built); lance-graph provides contract types OGAR consumes + query. graphrag is query-side ONLY, reads the calcified substrate, never ingests. A lance-graph doc-transcode crate is REJECTED (duplicates ogar-from-docv1, inverts the dep). - REUSE the substrate, BUILD only 3 gaps: hierarchical Leiden, HippoRAG-PPR reset-distribution, BM25. Extraction/SPO/fact-store/vector-CAM-PQ already exist; causal_edge::CausalEdge64 is the SPO edge (not the thinking-engine variant); EpisodicWitness64 is a queued column (today WitnessTable + EpisodicEdges64). graphrag-rs = REUSE-AS-REFERENCE (E-V3-GRAPHRAG-INV-1). - Retrieval = rung ascent (#708): dispatches through RungElevator + canonical RungLevel; BLOCK ascends and widens the CAUSES..BECOMES predicate-plane mask, FLOW relaxes to base. The graph is load-bearing because surprise ascends the elevator. Maps onto the triangle-tenants Maslow pyramid (D-TRI). - Probe-first: G0 P-GRAPH-LOADBEARING (graph vs vector-only) gates all Leiden/PPR code. Composes atop oxigraph-arigraph-cognitive-shader-soa-merge-v1. No bytes land; D-GR-5 (OGAR seam) mint-gated + baton-audited. Board hygiene: INTEGRATION_PLANS prepend in the same commit. Co-Authored-By: Claude <noreply@anthropic.com>
AdaWorldAPI
pushed a commit
that referenced
this pull request
Jul 17, 2026
…ate (§10) Verified every reuse/build assumption against the tree at 8d3209c (#708 merged). Verdict: FEASIBLE, two small non-blocking caveats, no circular deps. Confirmed reachable (pub): RungElevator (pub struct/fields + new + on_gate) and RungLevel/GateDecision/MailboxSoaView + CausalEdge64 all from ZERO-DEP crates; TripletGraph exposes the exact walk (get_associated multi-hop, intervene_on Pearl-2 apex, detect_contradictions = the BLOCK trigger, revise_with_evidence), spo_bridge::promote_to_spo pub, hdr_pagerank + ScentCsr::spmv pub, cam_pq pub. The rung loop maps 1:1 onto existing methods. Caveat 1 (drives a feature gate): the graph legs live in lance-graph core = drags datafusion/lance/arrow (19 heavy-dep lines); the rung-ascent core is zero-dep. => default graphrag = LIGHT (rung ascent + BM25 + in-memory PPR + G0 probe); feature graphrag-lance = HEAVY graph legs. §4 crate topology updated. Caveat 2: rung_widened_layer_mask is private (driver.rs:701) => make pub / move to contract, or replicate the pure fn (secondary leg; ascent works without it). Genuine BUILD confirmed: hierarchical Leiden (jc louvain is example-only), HippoRAG-PPR reset-distribution atop hdr_pagerank, BM25. Next step needs no upstream change: scaffold the light crate + cargo-check compile-proof, then G0. Co-Authored-By: Claude <noreply@anthropic.com>
AdaWorldAPI
pushed a commit
that referenced
this pull request
Jul 17, 2026
…lone crate Operator-directed architecture correction. Instead of a query-side crates/graphrag, expand AriGraph, which already owns the retrieval brain + episodic-witness basins. Grounding (verified against the tree): AriGraph already has retrieval.rs (OsintRetriever = BFS+episodic fusion), witness_corpus.rs, episodic.rs, spo_bridge.rs, markov_soa.rs — and has NO community/basin/cluster/Leiden/partition type (grep-confirmed). So Leiden fills the one structural-partition gap that COMPLEMENTS the episodic-witness partition AriGraph owns; it does not clutter or duplicate. A separate crate is a free-function-on-carrier's-state (the CLAUDE.md litmus reject) and a parallel retrieval layer beside an existing one. v1.1 topology: - arigraph/community.rs (NEW): hierarchical Leiden over the TripletGraph adjacency (triplets + entity_index; backed by in-core blasgraph CSR). - arigraph/retrieval.rs (EXTEND): + PPR/HippoRAG atop blasgraph::hdr_pagerank; community becomes a THIRD fusion signal beside BFS + episodic; driven by the #708 RungElevator (detect_contradictions -> BLOCK -> wider community/PPR walk). - DocGraphQuery trait stays in lance-graph-contract (impl = AriGraph methods). - BM25/lexical stays OUT of AriGraph (text-index, not a fact graph). - NO new crate. This dissolves the §10 dep-weight feature-gate (the graph capabilities live where the graph lives — core). New design constraint: no-singleton + write-on-behalf — community detection reads the graph the mailbox owns; a persisted membership is a born-stamped value-tenant lane. The #708 rung-ascent alignment, the reuse census, the OGAR boundary, and the probes all carry over unchanged. Board hygiene: INTEGRATION_PLANS entry annotated with the v1.1 scope. Co-Authored-By: Claude <noreply@anthropic.com>
AdaWorldAPI
pushed a commit
that referenced
this pull request
Jul 17, 2026
Evaluation + wave plan for feeding a new query-side crates/graphrag and OGAR's ogar-doc from the existing SoA + CausalEdge64 + (queued) EpisodicWitness64 substrate, aligned to #708 / D-TRI-6 (the rung-ascent loop) and the v3 waves. Key rulings the plan encodes: - Direction reframe (assembler-vs-storage boundary): OGAR owns doc ingestion + the document 0x080B mint (ogar-from-docv1 -> ogar-doc-ir -> ogar-doc, built); lance-graph provides contract types OGAR consumes + query. graphrag is query-side ONLY, reads the calcified substrate, never ingests. A lance-graph doc-transcode crate is REJECTED (duplicates ogar-from-docv1, inverts the dep). - REUSE the substrate, BUILD only 3 gaps: hierarchical Leiden, HippoRAG-PPR reset-distribution, BM25. Extraction/SPO/fact-store/vector-CAM-PQ already exist; causal_edge::CausalEdge64 is the SPO edge (not the thinking-engine variant); EpisodicWitness64 is a queued column (today WitnessTable + EpisodicEdges64). graphrag-rs = REUSE-AS-REFERENCE (E-V3-GRAPHRAG-INV-1). - Retrieval = rung ascent (#708): dispatches through RungElevator + canonical RungLevel; BLOCK ascends and widens the CAUSES..BECOMES predicate-plane mask, FLOW relaxes to base. The graph is load-bearing because surprise ascends the elevator. Maps onto the triangle-tenants Maslow pyramid (D-TRI). - Probe-first: G0 P-GRAPH-LOADBEARING (graph vs vector-only) gates all Leiden/PPR code. Composes atop oxigraph-arigraph-cognitive-shader-soa-merge-v1. No bytes land; D-GR-5 (OGAR seam) mint-gated + baton-audited. Board hygiene: INTEGRATION_PLANS prepend in the same commit. Co-Authored-By: Claude <noreply@anthropic.com>
AdaWorldAPI
pushed a commit
that referenced
this pull request
Jul 17, 2026
…ate (§10) Verified every reuse/build assumption against the tree at 8d3209c (#708 merged). Verdict: FEASIBLE, two small non-blocking caveats, no circular deps. Confirmed reachable (pub): RungElevator (pub struct/fields + new + on_gate) and RungLevel/GateDecision/MailboxSoaView + CausalEdge64 all from ZERO-DEP crates; TripletGraph exposes the exact walk (get_associated multi-hop, intervene_on Pearl-2 apex, detect_contradictions = the BLOCK trigger, revise_with_evidence), spo_bridge::promote_to_spo pub, hdr_pagerank + ScentCsr::spmv pub, cam_pq pub. The rung loop maps 1:1 onto existing methods. Caveat 1 (drives a feature gate): the graph legs live in lance-graph core = drags datafusion/lance/arrow (19 heavy-dep lines); the rung-ascent core is zero-dep. => default graphrag = LIGHT (rung ascent + BM25 + in-memory PPR + G0 probe); feature graphrag-lance = HEAVY graph legs. §4 crate topology updated. Caveat 2: rung_widened_layer_mask is private (driver.rs:701) => make pub / move to contract, or replicate the pure fn (secondary leg; ascent works without it). Genuine BUILD confirmed: hierarchical Leiden (jc louvain is example-only), HippoRAG-PPR reset-distribution atop hdr_pagerank, BM25. Next step needs no upstream change: scaffold the light crate + cargo-check compile-proof, then G0. Co-Authored-By: Claude <noreply@anthropic.com>
AdaWorldAPI
pushed a commit
that referenced
this pull request
Jul 17, 2026
…lone crate Operator-directed architecture correction. Instead of a query-side crates/graphrag, expand AriGraph, which already owns the retrieval brain + episodic-witness basins. Grounding (verified against the tree): AriGraph already has retrieval.rs (OsintRetriever = BFS+episodic fusion), witness_corpus.rs, episodic.rs, spo_bridge.rs, markov_soa.rs — and has NO community/basin/cluster/Leiden/partition type (grep-confirmed). So Leiden fills the one structural-partition gap that COMPLEMENTS the episodic-witness partition AriGraph owns; it does not clutter or duplicate. A separate crate is a free-function-on-carrier's-state (the CLAUDE.md litmus reject) and a parallel retrieval layer beside an existing one. v1.1 topology: - arigraph/community.rs (NEW): hierarchical Leiden over the TripletGraph adjacency (triplets + entity_index; backed by in-core blasgraph CSR). - arigraph/retrieval.rs (EXTEND): + PPR/HippoRAG atop blasgraph::hdr_pagerank; community becomes a THIRD fusion signal beside BFS + episodic; driven by the #708 RungElevator (detect_contradictions -> BLOCK -> wider community/PPR walk). - DocGraphQuery trait stays in lance-graph-contract (impl = AriGraph methods). - BM25/lexical stays OUT of AriGraph (text-index, not a fact graph). - NO new crate. This dissolves the §10 dep-weight feature-gate (the graph capabilities live where the graph lives — core). New design constraint: no-singleton + write-on-behalf — community detection reads the graph the mailbox owns; a persisted membership is a born-stamped value-tenant lane. The #708 rung-ascent alignment, the reuse census, the OGAR boundary, and the probes all carry over unchanged. Board hygiene: INTEGRATION_PLANS entry annotated with the v1.1 scope. Co-Authored-By: Claude <noreply@anthropic.com>
AdaWorldAPI
pushed a commit
that referenced
this pull request
Jul 17, 2026
Evaluation + wave plan for feeding a new query-side crates/graphrag and OGAR's ogar-doc from the existing SoA + CausalEdge64 + (queued) EpisodicWitness64 substrate, aligned to #708 / D-TRI-6 (the rung-ascent loop) and the v3 waves. Key rulings the plan encodes: - Direction reframe (assembler-vs-storage boundary): OGAR owns doc ingestion + the document 0x080B mint (ogar-from-docv1 -> ogar-doc-ir -> ogar-doc, built); lance-graph provides contract types OGAR consumes + query. graphrag is query-side ONLY, reads the calcified substrate, never ingests. A lance-graph doc-transcode crate is REJECTED (duplicates ogar-from-docv1, inverts the dep). - REUSE the substrate, BUILD only 3 gaps: hierarchical Leiden, HippoRAG-PPR reset-distribution, BM25. Extraction/SPO/fact-store/vector-CAM-PQ already exist; causal_edge::CausalEdge64 is the SPO edge (not the thinking-engine variant); EpisodicWitness64 is a queued column (today WitnessTable + EpisodicEdges64). graphrag-rs = REUSE-AS-REFERENCE (E-V3-GRAPHRAG-INV-1). - Retrieval = rung ascent (#708): dispatches through RungElevator + canonical RungLevel; BLOCK ascends and widens the CAUSES..BECOMES predicate-plane mask, FLOW relaxes to base. The graph is load-bearing because surprise ascends the elevator. Maps onto the triangle-tenants Maslow pyramid (D-TRI). - Probe-first: G0 P-GRAPH-LOADBEARING (graph vs vector-only) gates all Leiden/PPR code. Composes atop oxigraph-arigraph-cognitive-shader-soa-merge-v1. No bytes land; D-GR-5 (OGAR seam) mint-gated + baton-audited. Board hygiene: INTEGRATION_PLANS prepend in the same commit. Co-Authored-By: Claude <noreply@anthropic.com>
AdaWorldAPI
pushed a commit
that referenced
this pull request
Jul 17, 2026
…ate (§10) Verified every reuse/build assumption against the tree at 8d3209c (#708 merged). Verdict: FEASIBLE, two small non-blocking caveats, no circular deps. Confirmed reachable (pub): RungElevator (pub struct/fields + new + on_gate) and RungLevel/GateDecision/MailboxSoaView + CausalEdge64 all from ZERO-DEP crates; TripletGraph exposes the exact walk (get_associated multi-hop, intervene_on Pearl-2 apex, detect_contradictions = the BLOCK trigger, revise_with_evidence), spo_bridge::promote_to_spo pub, hdr_pagerank + ScentCsr::spmv pub, cam_pq pub. The rung loop maps 1:1 onto existing methods. Caveat 1 (drives a feature gate): the graph legs live in lance-graph core = drags datafusion/lance/arrow (19 heavy-dep lines); the rung-ascent core is zero-dep. => default graphrag = LIGHT (rung ascent + BM25 + in-memory PPR + G0 probe); feature graphrag-lance = HEAVY graph legs. §4 crate topology updated. Caveat 2: rung_widened_layer_mask is private (driver.rs:701) => make pub / move to contract, or replicate the pure fn (secondary leg; ascent works without it). Genuine BUILD confirmed: hierarchical Leiden (jc louvain is example-only), HippoRAG-PPR reset-distribution atop hdr_pagerank, BM25. Next step needs no upstream change: scaffold the light crate + cargo-check compile-proof, then G0. Co-Authored-By: Claude <noreply@anthropic.com>
AdaWorldAPI
pushed a commit
that referenced
this pull request
Jul 17, 2026
…lone crate Operator-directed architecture correction. Instead of a query-side crates/graphrag, expand AriGraph, which already owns the retrieval brain + episodic-witness basins. Grounding (verified against the tree): AriGraph already has retrieval.rs (OsintRetriever = BFS+episodic fusion), witness_corpus.rs, episodic.rs, spo_bridge.rs, markov_soa.rs — and has NO community/basin/cluster/Leiden/partition type (grep-confirmed). So Leiden fills the one structural-partition gap that COMPLEMENTS the episodic-witness partition AriGraph owns; it does not clutter or duplicate. A separate crate is a free-function-on-carrier's-state (the CLAUDE.md litmus reject) and a parallel retrieval layer beside an existing one. v1.1 topology: - arigraph/community.rs (NEW): hierarchical Leiden over the TripletGraph adjacency (triplets + entity_index; backed by in-core blasgraph CSR). - arigraph/retrieval.rs (EXTEND): + PPR/HippoRAG atop blasgraph::hdr_pagerank; community becomes a THIRD fusion signal beside BFS + episodic; driven by the #708 RungElevator (detect_contradictions -> BLOCK -> wider community/PPR walk). - DocGraphQuery trait stays in lance-graph-contract (impl = AriGraph methods). - BM25/lexical stays OUT of AriGraph (text-index, not a fact graph). - NO new crate. This dissolves the §10 dep-weight feature-gate (the graph capabilities live where the graph lives — core). New design constraint: no-singleton + write-on-behalf — community detection reads the graph the mailbox owns; a persisted membership is a born-stamped value-tenant lane. The #708 rung-ascent alignment, the reuse census, the OGAR boundary, and the probes all carry over unchanged. Board hygiene: INTEGRATION_PLANS entry annotated with the v1.1 scope. Co-Authored-By: Claude <noreply@anthropic.com>
AdaWorldAPI
pushed a commit
that referenced
this pull request
Jul 17, 2026
…arness Ship the buildable, ungated graphrag deliverables (plan graphrag-doc-retrieval-soa-integration-v1 §5). All pure / reversible / no-write-path capabilities land ahead of G0 exactly as D-GR-3a (#714); the D-GR-2 retrieval wiring stays gated on the G0 real-corpus verdict. D-GR-1 (contract, zero-dep): - doc_graph::{DocGraphQuery, ScoredId} - rung-aware document-graph read surface with a provided retrieve(seeds, RungLevel, top_k) default carrying the rung->walk dispatch (0-1 ranking / 2 SPO-G hop / 3+ wider community-scoped walk); the D-GR-2 design (OsintRetriever <-> #708 RungElevator) lives in the module-doc. 9 tests. D-GR-3b (lance-graph core, AriGraph): - arigraph::ppr - TripletGraph::personalized_pagerank (HippoRAG spread, confidence-weighted, deterministic, unit-sum). 6 tests. - arigraph::community - Leiden refine_connected (splits internally- disconnected Louvain communities into connected components). +2 tests. - arigraph::bm25 - Okapi BM25 lexical leg (Bm25Index, k1=1.2/b=0.75). 5 tests. G0: - examples/g0_graph_loadbearing.rs - the P-GRAPH-LOADBEARING harness: vector-only (BM25) vs vector+PPR+community over a synthetic multi-hop fixture, prints the with-vs-without delta. Scaffold, not the verdict - the real KILL/PASS needs a labeled corpus + jc::reliability. Plan sharpened per operator (§3a/§3b): (a) the cosine-replacement distance ALREADY EXISTS (certified bgz-tensor::fisher_z::{FamilyGamma, Base17Fz} rho>=0.999, clean-room helix) - wire it, retire the blasgraph popcount dead-code; do not rebuild. (b) part_of:is_a category = Leiden community = episodic-witness basin - one concept; P-COMMUNITY-BASIN-AGREE tests the identity. Verified: lance-graph 945 lib + 18 D-GR module tests; lance-graph-contract 10 tests; clippy -D warnings clean on new files; fmt clean. Board: EPIPHANIES E-GRAPHRAG-DGR3B-1, STATUS_BOARD graphrag section, LATEST_STATE contract inventory, AGENT_LOG, plan v1.2. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_016b33swuXE23hKtqxsHu9p1
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What
Executes the §3a Maslow-pyramid wiring that
E-MASLOW-PYRAMID-OF-COGNITION-1named as "the missing wiring" — the first code plateau of the rung/ancestry arc. Four deliverables, all verified green.(D) The rung ascent loop (centerpiece). The driver already held a per-driver
RungElevatorand calledon_gate()once per dispatch, but the rung never reached projection selection — stage [3] cascade usedreq.layer_maskverbatim. Now the elevator's current level (advanced by the previous cycle's gate) selects this cycle's cascade breadth viarung_widened_layer_mask, closing the loop across cycles.The one real correctness gate was an axis mismatch, not the ladder: p64
CognitiveShader::cascade'slayer_maskis an 8-bit predicate-plane mask (CAUSES..BECOMES), NOT the 3-bit SPO projection maskRungLevel::causal_mask_bits()returns. Feedingcausal_mask_bits()raw would be a category error and would collapse base-Surface to plane 0. Fix: the rung's Pearl level selects a predicate-plane widen-set (mirrorsp64_bridge::edge_to_layer_mask), applied as a UNION gated on elevation-above-base → identity at base (zero regression), superset-monotone above. The D-TRI-6 settlement probe (sustained BLOCK ascends → sustained FLOW relaxes to base; mask widens then returns to identity) is green.(A) RungLevel dedup + dead-path removal. thinking-engine's
RungLevelwas variant-identical to the contract's; collapsed topub use lance_graph_contract::cognitive_shader::RungLevel(addedas_u8()to the contract; a discriminant-parity test is the "measure" step — arepr(u8)variant-identical enum needs no jc battery). Deleted the deadshould_elevate,LayerId(zero consumers), andAgent.current_rung(write-once Surface, never mutated — the loop lives on the driver's elevator, not the orphaned Agent).(B)
scenario.rsPearl-numbering doc wobble fixed (interventions = rung 2, not 3). (C)nars_enginedeprecation silenced with a justified#[allow(deprecated)](v2inference_type()already routes throughfrom_mantissainternally;temporal()inert under v2) — plannerclippy -D warningsexit 0.Verification
†
-p cognitive-shader-driver -- -D warningsis blocked only by pre-existinglance-graph-ontologydep debt (TD-ONTOLOGY-CLIPPY-DEBT-1); the driver's own code has zero findings.‡ thinking-engine is a workspace-excluded (non-CI) crate carrying ~40 pre-existing clippy lints + a stale test-compile break, all outside P3's diff (
TD-THINKING-ENGINE-EXCLUDED-DEBT-1). Its lib builds clean.The System-1 arm (
apply_delta(rung_delta())) is left test-only — the driver hot path has no emergence/coherence scalar to feed it, and fabricating one is forbidden.Board hygiene (same commit)
STATUS_BOARD D-TRI-6 (ascent loop WIRED), EPIPHANIES
E-RUNG-ASCENT-WIRED-1, TECH_DEBTTD-THINKING-ENGINE-EXCLUDED-DEBT-1+TD-ONTOLOGY-CLIPPY-DEBT-1.🤖 Generated with Claude Code
https://claude.ai/code/session_01Awg6TXocHcwTtc6eGsHcdD
Generated by Claude Code
Summary by CodeRabbit
New Features
Documentation
Refactor