NEW infectious disease dossier — no manifest / atoms / package on disk yet; manifest field is intentionally blank at PLANNED (prisma/seed/manifests/id.influenza.core.v1.{ts,atoms.ts} are out-of-shard scope per §5.5 pragmatic policy per peds.febrile-infant 0815a900 precedent — will be authored in a future shard once the seed manifest ships). NEXT STEPS (for SCAFFOLDED promotion): (1) author manifest at prisma/seed/manifests/id.influenza.core.v1.ts; (2) build antiviral decision atoms (oseltamivir vs baloxavir by severity/timing); (3) add bacterial superinfection detection atoms; (4) integrate ARDS protocol handoff for critical influenza pneumonia; (5) register influenza-specific diagnostic panel (rapid molecular + RT-PCR) and antiviral-resistance workup in clinical-tools-registry. Key clinical decision point: antiviral within 48h of symptom onset — benefit diminishes after; however, hospitalized and immunocompromised patients benefit even if >48h, AND high-risk-for-complications outpatients (pregnancy / age ≥ 65 / chronic comorbidity / immunocompromised / morbid obesity) benefit regardless of strict 48-h window per IDSA 2018. Baloxavir single-dose convenience vs oseltamivir BID × 5d — patient preference and compliance factor; baloxavir contraindicated in pregnancy (CDC ACIP) and not recommended in severely immunocompromised (limited efficacy data per IDSA 2018). Bacterial superinfection (especially MRSA) is major cause of influenza mortality — biphasic illness pattern is classic and must be caught early (IDSA 2018); cross-dossier routing to pulm.cap.core.v1 owns the bacterial CAP management; id.sepsis.core.v1 owns the SSC bundle if septic shock develops. Pregnancy is independent high-risk factor — oseltamivir preferred, empiric treatment without waiting for test results (IDSA 2018; CDC ACIP). Emerging WHO 2024 signal: combination oseltamivir + baloxavir in severe immunocompromised with prolonged shedding and progressive disease — NOT yet routine IDSA standard; case-series + cohort supportive; ID consultation drives individualized decision; resistance genotyping (H275Y NA mutation for oseltamivir; PA I38T for baloxavir) if no improvement by day 3-5. Deepened 2026-05-14 (shard-5-obped-id depth-pass-1): added co-located _briefs/id.influenza.core.v1.md (design brief — frame, 5-axis phenotype matrix [strain × severity tier × host-risk × complication × antiviral-window], settings axis prehospital → outpatient → ED → ward → ICU → outpatient follow-up, severity-triggers cross-walk, Bayesian linkage summary, 12-phase author order, Stage-A API checklist, open-gaps log, schema proposal cache) + _research-bundles/id.influenza.core.v1.md (PMIDs, decision-threshold table per scenario, LR table per finding, conditional dependencies, cross-dossier Bayesian routing, gap log). Closed audit-FAIL by: blanking manifest field at PLANNED (atoms field removed entirely; both files are out-of-shard scope); repointing workup.bacterial_superinfection → workup.hap_vap (canonical IDSA 2016 HAP/VAP registry entry which covers anti-MRSA + anti-pseudomonal empiric therapy + routes to pulm.cap.core.v1 / id.sepsis.core.v1); repointing workup.ards_critical → workup.ards (canonical ARDS registry entry routing to pulm.ards.core.v1); repointing workup.myocarditis → workup.pericarditis_myocarditis (canonical registry entry); repointing panel.cmp → panel.renal (canonical BMP/renal analyser); repointing panel.procalcitonin → panel.inflammation (canonical CRP+PCT panel); removing workup.rapid_molecular_influenza, workup.rt_pcr_influenza, workup.cxr_pneumonia, workup.antiviral_resistance, panel.influenza_rapid_molecular, panel.influenza_rt_pcr, panel.blood_cultures (none resolve in clinical-tools-registry today; testing + resistance genotyping documented inline in flow.required_inputs + flow.phases + severity_triggers; registry additions deferred to a future shard with seed manifest authorship). Added 5 new severity triggers (high_risk_outpatient_without_antiviral severe — start oseltamivir even if > 48 h symptoms in high-risk per IDSA 2018; bacterial_superinfection_features severe — anti-MRSA + anti-pseudomonal coverage + cross-dossier route to pulm.cap.core.v1 / id.sepsis.core.v1; ards_progression life_threatening — ICU + ARDSnet lung-protective vent + prone if PaO2/FiO2 < 150 + VV-ECMO consideration + cross-dossier route to pulm.ards.core.v1; influenza_encephalitis_features severe — neuro consult + empiric IV acyclovir pending HSV PCR + MRI brain + continue oseltamivir; influenza_myocarditis_features severe — echo + cardiology consult + telemetry + continue oseltamivir + consider CMR). Preserved all 4 existing severity triggers. Bumped evidence.last_reconciled to 2026-05-14. Promoted PLANNED→INTEGRATED 2026-05-22 (shard-5 build campaign): manifest repointed to the id.sepsis sibling precedent; all PMIDs live-verified — the prior 6 placeholders were mis-attributed (the author had flagged 30517394 as suspect for IDSA-Uyeki; PubMed confirms 30517394 is a nursing-education article and 30566567 is the real Uyeki 2018 IDSA guideline), corrected to IDSA 30566567 + Dobson oseltamivir meta 25640810 + baloxavir CAPSTONE-1 30184455 (confirmed real). Antiviral RxCUIs corrected: oseltamivir 283150→260101, baloxavir marboxil 2173025→2099995, peramivir 1315411→619693 (all prior codes invalid). Phenotype matrix (strain × severity tier × host-risk × complication × antiviral-window — 5-axis 120-cell cross-product collapsed by clinical meaning) encoded indirectly via severity_triggers (critical_influenza_pneumonia, bacterial_superinfection, influenza_in_pregnancy, immunocompromised_prolonged_shedding, high_risk_outpatient_without_antiviral, bacterial_superinfection_features, ards_progression, influenza_encephalitis_features, influenza_myocarditis_features) + per-setting playbook logic (outpatient, inpatient) + regimen_axes (uncomplicated outpatient, hospitalized severe, pregnancy, post-exposure prophylaxis) + flow phase logic (FRAME / ENTRY / CONTEXT / RED_FLAGS / INITIAL_WORKUP / BRANCHING_WORKUP / DIFFERENTIAL / RISK_STRATIFICATION / TREATMENT / DISPOSITION / MONITORING / FOLLOWUP). First-class TS field for phenotype matrix is schema-blocked. Bayesian linkage (pre-test priors: ILI during peak season with documented community activity ≥ baseline → influenza ~ 30-60%; off-season ~ 1-5%; hospitalized respiratory illness during season ~ 10-30%; bacterial superinfection in confirmed influenza ~ 10-15%; ARDS in hospitalized influenza ~ 5-10%; myocarditis ~ 0.5-2%; encephalitis ~ 0.1-0.5%. LRs per finding: rapid molecular assay LR+ ~ 95-100 / LR− ~ 0.01-0.05; RIDT LR+ ~ 5-14 / LR− ~ 0.3-0.55 — negative RIDT does NOT exclude per IDSA 2018; RT-PCR LR+ > 100 / LR− < 0.01; CXR focal consolidation for superinfection LR+ ~ 4-6; procalcitonin > 0.25 for superinfection LR+ ~ 3-5; biphasic fever + new productive cough for superinfection LR+ ~ 3-4. T_treat (empiric antiviral) ~ post-test influenza ≥ 30% OR ANY high-risk-for-complications + ILI OR ANY hospitalized + ILI during season; T_test ~ post-test influenza ≥ 10% + low-risk outpatient; T_combo ~ severe immunocompromised + progressive despite monotherapy + WHO 2024 emerging recommendation. Cross-dossier routing: pulm.cap.core.v1 for bacterial superinfection confirmed; id.sepsis.core.v1 for septic shock; pulm.ards.core.v1 for ARDS PaO2/FiO2 < 200; neph.aki.core.v1 if AKI develops; prev.adult-immunization.core.v1 for post-illness vaccination reconciliation) documented in _research-bundles/id.influenza.core.v1.md. ROS/DDx LR seed data audited by npm run audit:ros-ddx-coverage (cross-cutting; not touched by this shard). Prehospital recognition is currently encoded implicitly via existing entry_points (ili_syndrome, hospitalized_influenza) + outpatient setting_playbook; a first-class "prehospital" DossierSetting value is schema-blocked.