DEM-PRODUCER-HTF (reactor): real higher-timeframe bias; structure passes its 0.40 cap - #91
Merged
Merged
Conversation
…act; goldens regenerated once, itemized DEM-GOV §9 DEM-PRODUCER-PLAN (owner-authorized 2026-08-25; afi-governance #55; §9 determinations D-2/D-4/D-5 in #56). Merge AFTER afi-core dem/plan-core and afi-config dem/plan-config. Source: - uss/cpjMapper.ts: the submitted plan survives as uss.plan (afi.trade-plan.v1, decimal strings, validated against the governed contract) or is refused at ingest (typed error → 422 trade_plan_invalid). TV/MarkitTick untouched. - enrichment/tradePlanVerification.ts (NEW): the D-DEM-5(6) law — every submitted level within [L − W, H + W] of the fetched window; plan geometry from prices only (D-DIR-3 by construction); R:R to the nearest target, 4 dp; refusal = TradePlanVerificationError (NodeConfigurationError). - providers/adapters/technicalLocalAdapter.ts: the producer — verifies the plan against ITS fetched candles and emits technical.plan (declared absences: no plan → no block; no stop/target → no R:R). - pipeline/nodes/laneView.ts: viewTechnical projects plan verbatim. - server.ts: 422 trade_plan_unverifiable (unwrapping NodeExecutionError.cause), 422 trade_plan_invalid; both persisted:false. - types/TradePlan.ts (NEW), types/UssLenses.ts, uss/ussValidator.ts: types. Registries/pins: the three fixture trees mirror afi-config (mapping 1.1.0, config, three implementationVersion moves, registration hash 5cb9b7a4…); registryLoader/officialArtifactPins pins rotated (pluginSetHash 36f911f4…). Goldens: the two plan-bearing CPJ fixtures re-authored inside the demo envelope (old levels were unverifiable by the law this slot lands); 12 goldens regenerated ONCE; every moved path itemized in INTENTIONAL_DIFFS.md (risk 0.9 → {0.5 via the provider's rr 1.4286 | 0.2 via the declared floor}; executionSummaryHash flips executed → degraded on the two goldens where the floor is the first fired default; identity-hash class on all 12). test/oracle/support/goldenDiff.mjs (NEW): the committed per-path differ. Tests: verification KATs (band, geometry, absences, error class), carrier (decimal strings, refusals), producer (role split, D-DIR-3 negative test, refusal, kernel floor, envelope-covers-fixtures), error-table rows for both 422s (no record), seam suites re-oracled to the composition reference with the mapping resolved by mappingRef from the fixture registry, golden scorerInput captured at the rubric call. 756/756; typecheck; compiled-build proofs (unavailable 503; Mongo persistence 10/10; oracle equivalence 7/7; SIGTERM) green locally. No scoring-law value moves. Sealed records untouched. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
…n law An adversarial review of the branch (7 lenses, findings independently verified) found five real defects in the D-DEM-5(6) law. All five fixed here; no golden byte moves (every fix is refusal-side or message-side). - MAJOR: an entry RANGE was verified only at its conservative bound, so a range containing (or on the wrong side of) its own stop was ACCEPTED and an R:R emitted from one end. The stop must now clear the FAR bound as well: long -> stop < entryLow, short -> stop > entryHigh. Three refusal KATs (long straddle, short straddle, far bound EQUALS the stop) + the accept-side edge. - MINOR: the inclusive band edge was false on non-integer windows (binary float: 1.1 - 0.2 = 0.9000000000000001 refused an entry of exactly "0.9"). Relative tolerance (1e-12) at both edges + bounded-precision numbers in the refusal message; a non-integer-window KAT pins both edges. - MINOR: a non-array takeProfits (or a non-object levels) threw a bare TypeError, escaping the 422 unwrap. Both refuse properly now, with KATs. - MINOR: "half-up to 4 decimals" was not what Math.round(v*1e4)/1e4 does at binary ties. The law is documented as what it is (deterministic binary quantisation) and the unfavourable tie (0.00015 -> 0.0001) is pinned. - MINOR: missing edge KATs added: empty takeProfits array (no-target case), degenerate entry range (min === max), exotic numeric strings, an overflow-to-Infinity decimal string, malformed candles; the hedged short-geometry regex is tightened to its exact message. 756 -> 764 tests, all green; typecheck green; goldens byte-unchanged. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
DEM-GOV §9 DEM-PRODUCER-CANDLE (owner-authorized 2026-08-25; afi-governance #55; determinations D-2/D-3 in #56) — the act D5-GOV D-D5-1 reserved. Merge AFTER afi-core dem/candle-core and afi-config dem/candle-config, which merge after the whole PLAN wave. Source: - enrichment/candleStructure.ts (NEW): the two producers as pure functions over the window the lane already fetches. brokeEmaWithBody = the latest bar's body closed on the counter-trend side of EMA20 (bullish: close < ema20; bearish: close > ema20; range: the body crossed EMA20 this bar) — colour-free, one law per side. Heikin-Ashi flat-back = the recurrence over the whole window, seed (o+c)/2, epsilon 0 (bullish: an up HA bar with low >= haOpen; bearish: a down HA bar with high <= haOpen), confirmed iff its side agrees with the lane's own EMA trend law. Both read the lane's computed trendBias, never a submitted direction (DIR-GOV D-DIR-3). - enrichment/technicalIndicators.ts: emits the three facts with the bundle. - pipeline/nodes/laneView.ts: viewTechnical projects them; the imported stub constant and its pin are DELETED. - types/UssLenses.ts: lens payload widened. Registries/pins: three fixture trees mirror afi-config (mapping 1.2.0, config, three implementationVersion moves); the two pin tests rotated. Goldens (regenerated ONCE, itemized in INTENTIONAL_DIFFS.md): the SOL 15m seed (tv-neutral) computes brokeEmaWithBody TRUE, so structure 0.4 -> 0.15 and risk 0.2 -> 0.0 there; the other ten are unmoved in scored values. THE GATE IS MET: structure is non-constant over the fixture corpus ({0.4, 0.15}) and so is risk ({0.5, 0.2, 0.0}). haFlatBackConfirmed is false on all twelve BY CONSTRUCTION (every demo seed is a range regime) — recorded honestly; its branches are proven by KATs, not by the corpus. Tests: candleStructure.test.ts (12 KATs: every brokeEma branch incl. the green-body-below and straddle cases, both flat-back sides, the exact-equality no-wick edge, the doji, seed-independence at the 50-bar floor, both producers together); fiveLaneAdapters' stub assertions rewritten to computed-fact assertions (and that the stub export is GONE); guardrail fixture views carry the required facts; the seam test's bare-technical probe updated (the candle binds are required, so a bare namespace now refuses). 764/764; typecheck green. No scoring-law value moves. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
…ses its 0.40 cap DEM-GOV §9 DEM-PRODUCER-HTF (owner-authorized 2026-08-25; afi-governance #55; determinations D-1/D-2 in #56) — the mission DIR-GOV D-DIR-3's scope-guard reserved. Merge AFTER afi-core dem/htf-core and afi-config dem/htf-config. Source: - providers/adapters/technicalLocalAdapter.ts: reads the REGISTERED composition value (nodeOverrides.technical.config.htf, boot-validated against the closed paramsSchema) and fetches the signal window + both HTF windows CONCURRENTLY (Promise.all — this lane is the entry node, so serial round-trips delay every downstream wave). Each HTF window runs the SAME EMA trend law; a window below the 50-candle floor emits no sub-block (declared absence). A malformed registered value refuses (NodeConfigurationError), never guesses. - pipeline/nodes/laneView.ts + types/UssLenses.ts: technical.htf projected. - test/support/deterministicPriceFeedAdapter.ts (TEST ONLY): a deterministic per-symbol drift for timeframes >= 1d, so the fixture corpus exercises aligned, conflicting and neutral HTF. Sub-daily timeframes byte-untouched. Goldens (regenerated ONCE, itemized in INTENTIONAL_DIFFS.md): THE GATE IS MET — the aligned-HTF +0.4 structure term fires for the first time, so structure reaches 0.8 on the four BTC goldens, BEYOND its former 0.40 cap, and spans {0.15, 0.4, 0.8} over the corpus. The ETH pair exercises a real higher-timeframe CONFLICT (weekly bearish, daily bullish): the analyst's verdict is "unknown" — the branch DIR-GOV reserved, now reachable, resolved in afi-core, and pinned by a golden. inputHash/manifestHash byte-equal everywhere. Tests: htfProducer.test.ts — the producer (registered timeframes actually drive the fetches; concurrency proven by an in-flight peak of 3; kernel-floor absence; malformed-config refusal) and THE DIR-GOV D-DIR-3 NEGATIVE TEST the gate requires: varying facts.direction over long/short/neutral leaves the entire technical payload byte-identical, the bias facts contradict the submitted side where the market does, and a comment-stripped source scan proves no producer or projection file reads the submitted direction (with the scan proven to bite, and the lawful record-stamping reader shown to be outside the scanned set). 785/785; typecheck green; all four compiled-build proofs green against a real replica set (503-unavailable, 10 persistence, 7 oracle-equivalence, SIGTERM). No scoring-law value moves. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
|
Kilo Code Review could not run — your account is out of credits. Add credits or switch to a free model to enable reviews on this change. |
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
1 participant
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.
Wave 3 of 4 —
DEM-PRODUCER-HTF, step 3 of 3. Merge after afi-coredem/htf-coreand afi-configdem/htf-config.The capability the technical lane lacked
The adapter now reads the registered composition value and fetches the signal window plus both higher-timeframe windows concurrently (
Promise.all— this lane is the pipeline's entry node, so serial round-trips would delay every downstream wave and eat into TradingView's 3-second cancel). Each HTF window runs the same EMA20/EMA50 trend law; a window below the 50-candle floor emits no sub-block — a declared producer absence, never a guessed bias. A malformed registered value refuses rather than guessing.The gate's headline: structure passes its former 0.40 cap
analystScore.directioncpj-blofin-perp-long.*,tv-long.*)cpj-coinbase-spot-sell.*,tv-short.*)*-neutral.*)The aligned-HTF +0.4 structure term fires for the first time in this system's history: structure reaches 0.8, beyond its former
0.40cap, and spans {0.15, 0.4, 0.8} over the corpus. The ETH pair exercises a genuine higher-timeframe conflict, so the"unknown"verdict — the branch DIR-GOV reserved — is now reachable, resolved, and pinned by a golden.DIR-GOV D-DIR-3 — the dedicated negative test the gate requires
test/pipeline/htfProducer.test.ts:facts.directionover long/short/neutral leaves the entire technical payload byte-identical;Also proven: the registered timeframes actually drive the fetches; concurrency (in-flight peak of 3); kernel-floor absence; malformed-config refusal.
Test-only fixture change, stated plainly
The deterministic test feed gains a per-symbol drift applied only to timeframes ≥ 1d, so the corpus exercises aligned, conflicting and neutral HTF. Sub-daily timeframes are byte-untouched — every golden's own-timeframe series, indicators and pattern facts are unchanged, and the differ confirms it.
785/785 green; typecheck green; all four compiled-build proofs green against a real replica set; every moved golden path itemized with an explicit byte-EQUAL list.
🤖 Generated with Claude Code