تشغيل أي مهارة في Manus بنقرة واحدة

heater-appb-lineup-rec-review

النجوم٢

التفرعات٠

آخر تحديث٦ يوليو ٢٠٢٦ في ١٩:٠١

Use when assessing whether a HEATER lineup recommendation is safe to enact — /api/lineup/optimize or /api/start-sit output, the Optimizer page's start/sit calls, daily swaps, streaming pickups — "does this lineup make sense?", "why is it benching X?", or before applying any roster moves it suggests.

التثبيت

التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.

تشغيل في Manus

المصدر

hicklax13

hicklax13/HEATER_v1.0.0

فتح مستودع GitHub عرض مستودعات المنشئ

تنزيل

تشغيل في Manus

المهن ذات الصلةSOC

استنادا إلى تصنيف SOC المهني

مطوّرو البرمجياتمهن الحاسوب والرياضيات·SOC 15-1252

SKILL.md

readonly

name	heater-appb-lineup-rec-review
description	Use when assessing whether a HEATER lineup recommendation is safe to enact — /api/lineup/optimize or /api/start-sit output, the Optimizer page's start/sit calls, daily swaps, streaming pickups — "does this lineup make sense?", "why is it benching X?", or before applying any roster moves it suggests.

Lineup/Start-Sit Recommendation Review

REQUIRED BACKGROUND: heater-appb-manager-rec-review (league truth, data gathering, verdict protocol). This is the lineup-specific checklist. Payload fields: LineupSlot{status, value 0-100, matchup, current_slot, forced_start, reason} + daily: DailyMeta{urgency, rate_modes, ip_pace, swaps} + bench/impact/fa_suggestions{urgency_categories} (api/contracts/lineup.py:24-91).

The checklist

Slot legality. Every started player eligible for the assigned slot (eligibility = Yahoo positions / roster_slot; assignment = selected_position/current_slot — do not confuse them: confusing them once routed SP,RP swingmen into the hitter table). Counts exactly C/1B/2B/3B/SS/3×OF/2×Util/2×SP/2×RP/4×P. No IL/DTD player in an active slot.
Probables gate. Every started pure SP must be a confirmed/probable starter for the scope date (check the probables board or statsapi, ET dates). The engine zeroes non-probable pure SPs (volume=0) — a started NOT_PROBABLE SP means the gate broke (it WAS dead code once; the DCV gate fix pinned it).
Locked games. Today-scope recs must not move players whose games already started (ET clock — src/et_today.py; UTC evenings falsely unlock). Locked rows should render "—"/non-actionable, not 0.00.
IP pace coherence. ip_pace vs floor 20 / target ≈53.85 (src/ip_tracker.py:16,20). A "DANGER — stream SP immediately" call must square with (a) actual remaining probables this week and (b) adds_remaining_this_week > 0. Failure story: live showed "0 IP DANGER" for a healthy staff — the pool's projected ip was 0 (FanGraphs-403/Marcel gap), fixed by the _ip_pace YTD fallback (lineup_service.py:~506-569). Check freshness before believing an extreme pace claim.
Rate-mode sanity. rate_modes (ERA/WHIP → protect|compete|abandon) must match the live matchup state: protect when clearly winning the cat, abandon only when it's mathematically gone. An "abandon" stance on a winnable cat once traced to zeroed pitching projections — data, not strategy.
Bench-a-stud test. For every benched player with strong consensus/recent production, the payload must contain the evidence: low DCV value, bad matchup (elite opposing offense, hitter park), not probable, locked, or IL. forced_start=True marks the engine admitting a bad start was roster-forced — surface those. No evidence = DO-NOT-ENACT + investigate.
Cross-surface agreement. Same player, same day: Optimizer lines/slots must agree with the Matchup tab (shared slot order — web/src/lib/slot-order.ts mirrors the backend; ordering by RECOMMENDED position instead of assigned slot was a live bug). Win-prob mentions must equal the seeded outcome_probs number.
Scope arithmetic. today = daily DCV; rest_of_week/rest_of_season = LP with window-scaled totals (counting ÷ 26 per week). A weekly total that looks like a season total (or vice versa) is the window-scale bug class.
Streaming/FA suggestions. For each fa_suggestions entry: live-verify the player is STILL an FA (never trust a cached list — standing rule); urgency_categories are genuinely losing/tied cats; the implied drop is legal (not rounds-1-3 can't-drop, not an IL stash the UI protects); budget respects the 10/week combined cap; two-start weeks valued with the 0.93 second-start fatigue, not double full-quality.
Recompute one number. E.g. proposed staff WHIP = Σ(bb+h)/Σip from the payload's own lines — catches mean-of-rates and volume-weighting regressions instantly.

Red flags (instant DO-NOT-ENACT pending investigation)

Empty matchup strings across the board (the statsapi home_name mapping bug class) · every value ≈ 100 (normalization saturation) · literal "nan" anywhere (reason NaN class, fixed at lineup_service.py:459-460) · an empty lineup with HTTP 200 (the pipeline.run() no-such-method incident shipped exactly this) · started SP facing an elite offense with low value and forced_start=False.

When NOT to apply

Judging the ALGORITHM (weights, DCV math) → heater-appb-fantasy-math-invariants + calibration-backtest. Yahoo write execution is out of scope (writes 401 by design — enactment is manual).

المزيد من هذا المستودع

نفس المستودع

heater-appb-architecture

hicklax13/HEATER_v1.0.0

Use when starting any task on HEATER's live product — the api/ FastAPI backend (Railway service "celebrated-respect") or web/ Next.js frontend (heater-v1-0-1.vercel.app) — or when unsure where code lives, how a request flows, which database owns what, why CORS/Yahoo-writes/billing seem "missing", or which app (Streamlit vs React) is the real one. Load this first on App B work.

2026-07-062

heater-appb-auth-billing

hicklax13/HEATER_v1.0.0

Use when working on HEATER App B auth or billing — Clerk JWT verification, unexpected 401/402, HEATER_API_WRITE_TOKEN, Stripe checkout/webhooks/subscription tiers, require_pro gating, tier-based AI spend caps — or when auth "works locally but not live" (or vice versa), or a page bounces to /sign-in on hard load.

2026-07-062

heater-appb-bubba-ai

hicklax13/HEATER_v1.0.0

Use when working on Bubba, HEATER's AI assistant — /api/chat/* endpoints, provider streaming (src/ai/providers.py), token metering and spend caps, AI provider keys, attachments/page-context, saved prompts, or the chat composer in web/src/components/bubba/.

2026-07-062

heater-appb-calibration-backtest

hicklax13/HEATER_v1.0.0

Use when proving a HEATER algorithm works — calibrating constants, backtesting projections or recommendations, validating win probabilities, adding uncertainty intervals, checking calibration drift — or before letting any new math drive user-facing recommendations. Also when asked "is this signal real?" or when a calibration run finds no clear winner.

2026-07-062

heater-appb-contracts-openapi

hicklax13/HEATER_v1.0.0

Use when editing api/contracts/*, when api/openapi.json or web/src/lib/api/generated.ts drift (test_openapi_contract or the CI openapi-ts-sync job failing), when JSON responses 500 or render "nan"/"+nan" badges, or when adding/renaming fields on a response the live frontend already consumes.

2026-07-062

heater-appb-debugging

hicklax13/HEATER_v1.0.0

Use when App B misbehaves and the cause isn't obvious — works locally but not on Railway/Vercel (or vice versa), a live page renders empty/zero/wrong data, an endpoint 500s or times out, or tests see different data than the app. Run this playbook before changing code.

2026-07-062

name	heater-appb-lineup-rec-review
description	Use when assessing whether a HEATER lineup recommendation is safe to enact — /api/lineup/optimize or /api/start-sit output, the Optimizer page's start/sit calls, daily swaps, streaming pickups — "does this lineup make sense?", "why is it benching X?", or before applying any roster moves it suggests.

Lineup/Start-Sit Recommendation Review

The checklist

Slot legality. Every started player eligible for the assigned slot (eligibility = Yahoo positions / roster_slot; assignment = selected_position/current_slot — do not confuse them: confusing them once routed SP,RP swingmen into the hitter table). Counts exactly C/1B/2B/3B/SS/3×OF/2×Util/2×SP/2×RP/4×P. No IL/DTD player in an active slot.
Probables gate. Every started pure SP must be a confirmed/probable starter for the scope date (check the probables board or statsapi, ET dates). The engine zeroes non-probable pure SPs (volume=0) — a started NOT_PROBABLE SP means the gate broke (it WAS dead code once; the DCV gate fix pinned it).
Locked games. Today-scope recs must not move players whose games already started (ET clock — src/et_today.py; UTC evenings falsely unlock). Locked rows should render "—"/non-actionable, not 0.00.
IP pace coherence. ip_pace vs floor 20 / target ≈53.85 (src/ip_tracker.py:16,20). A "DANGER — stream SP immediately" call must square with (a) actual remaining probables this week and (b) adds_remaining_this_week > 0. Failure story: live showed "0 IP DANGER" for a healthy staff — the pool's projected ip was 0 (FanGraphs-403/Marcel gap), fixed by the _ip_pace YTD fallback (lineup_service.py:~506-569). Check freshness before believing an extreme pace claim.
Rate-mode sanity. rate_modes (ERA/WHIP → protect|compete|abandon) must match the live matchup state: protect when clearly winning the cat, abandon only when it's mathematically gone. An "abandon" stance on a winnable cat once traced to zeroed pitching projections — data, not strategy.
Bench-a-stud test. For every benched player with strong consensus/recent production, the payload must contain the evidence: low DCV value, bad matchup (elite opposing offense, hitter park), not probable, locked, or IL. forced_start=True marks the engine admitting a bad start was roster-forced — surface those. No evidence = DO-NOT-ENACT + investigate.
Cross-surface agreement. Same player, same day: Optimizer lines/slots must agree with the Matchup tab (shared slot order — web/src/lib/slot-order.ts mirrors the backend; ordering by RECOMMENDED position instead of assigned slot was a live bug). Win-prob mentions must equal the seeded outcome_probs number.
Scope arithmetic. today = daily DCV; rest_of_week/rest_of_season = LP with window-scaled totals (counting ÷ 26 per week). A weekly total that looks like a season total (or vice versa) is the window-scale bug class.
Streaming/FA suggestions. For each fa_suggestions entry: live-verify the player is STILL an FA (never trust a cached list — standing rule); urgency_categories are genuinely losing/tied cats; the implied drop is legal (not rounds-1-3 can't-drop, not an IL stash the UI protects); budget respects the 10/week combined cap; two-start weeks valued with the 0.93 second-start fatigue, not double full-quality.
Recompute one number. E.g. proposed staff WHIP = Σ(bb+h)/Σip from the payload's own lines — catches mean-of-rates and volume-weighting regressions instantly.

Red flags (instant DO-NOT-ENACT pending investigation)

When NOT to apply

Judging the ALGORITHM (weights, DCV math) → heater-appb-fantasy-math-invariants + calibration-backtest. Yahoo write execution is out of scope (writes 401 by design — enactment is manual).