contract: TemporalPov range filter + knowledge/board: stacked-awareness & opponent-modeling arcs (waves 1–2)#686
Conversation
…ecorded (real-play green, synthetic-style fails named) Co-Authored-By: Claude <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Awg6TXocHcwTtc6eGsHcdD
…rror of the temporal.rs version-range/rung read Adds lance_graph_contract::temporal_pov with VersionRange (half-open [from, to) interval over LanceVersion) and TemporalPov (range + rung), mirroring lance-graph-planner::temporal's QueryReference/deinterlace version-range vocabulary (crates/lance-graph-planner/src/temporal.rs) without reimplementing EpistemicMode/TemporalStatus/classify/deinterlace, which stay downstream in the planner crate. TemporalPov::at(ref_version, rung) mirrors QueryReference::at; admits() covers only the version-range half of admission (EpistemicMode::Strict's documented "row_version <= ref_version" semantics), leaving per-row knowable_from/Spoiler classification to the planner's classify/deinterlace. Instantiates E-MARKOV-TEMPORAL-STREAM-1 (.claude/board/EPIPHANIES.md, 2026-07-10) and its measured worked example D-SF-EPISODIC-1 (.claude/knowledge/stockfish-nnue-as-perturbation-cascade.md). 11 unit tests + 1 doctest, cargo test/clippy -p lance-graph-contract clean. LATEST_STATE.md Contract Inventory updated same commit per board hygiene. Co-Authored-By: Claude <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Awg6TXocHcwTtc6eGsHcdD
…st partials, TemporalPov + lichess-rs recorded 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_6aec33f1-c065-4036-be1f-874b44b67518) |
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Organization UI Review profile: CHILL Plan: Pro Plus Run ID: 📒 Files selected for processing (6)
📝 WalkthroughWalkthroughAdds a public ChangesTemporal POV contract
Estimated code review effort: 3 (Moderate) | ~20 minutes Possibly related PRs
Poem
🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✨ Finishing Touches📝 Generate docstrings
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 |
What
Three commits:
eee259be— wave-1 synthesis (knowledge doc + board): the stacked-awareness/opponent-modeling arc — five stockfish-rs probes recorded with honest grades (awareness ladder L0→L1 ~100×; L0 is structurally the codex POV defect measured; naive + diversion-vector L2 both fail on the synthetic corpus; NNUE prices sacrifice compensation; wedge detection hits the pre-registered Nxb5; counterfactual rung 5/5; phase-mask filter hurts retrieval while phase proves emergent in accumulator geometry).E-SF-AWARENESS-OPPONENT-ARC-1prepended.0ed93b59— contract:TemporalPov(operator-directed): zero-deptemporal_povmodule inlance-graph-contract—LanceVersion, half-openVersionRange(contains/intersect/len),TemporalPov { range, rung }withat(ref_version, rung)mirroring the planner'sQueryReference::at. Deliberately does NOT duplicate the planner'sEpistemicMode/classify/deinterlace(avoids the documented type-duplication anti-pattern; doc-comments name the canonical source). 11 unit tests + 1 doctest; clippy-D warningsclean; LATEST_STATE Contract Inventory updated in the same commit.648a8402— wave-2 synthesis: trap-lure GREEN (bait rate 0.767, +1592 cp pooled, safety clause held; scope honest — ground-truth style models; the composite finding: lure synthesis works, per-opponent style inference is the bottleneck), holes/piece-palette/lichess honest partials with named follow-up knobs, the lichess-rs scaffold + baked-release/JIT game-DB design note, the standing statistical-toolbox rule (ndarray::hpc::reliability+crates/jcJirak pillar; per-player style stability as an ICC question), and queued D-SF-FILTER-1.E-SF-TRAP-LURE-GREEN-1prepended; AGENT_LOG entries for both waves.Companion PRs/artifacts: stockfish-rs PR #10 (the eight probe examples),
AdaWorldAPI/lichess-rsinitial scaffold (ce44adaon main).🤖 Generated with Claude Code
https://claude.ai/code/session_01Awg6TXocHcwTtc6eGsHcdD
Generated by Claude Code
Summary by CodeRabbit
New Features
Documentation
Tests