Skip to content
Back to Reports
WAR Room synthesis v1.0

Country Excellence WAR Room — 100 Experts × 100 Rounds

The public synthesis of SongForgeAI’s 100-expert, 100-round Country Excellence WAR Room — 15 builds (C1-C13, B2896-B2909) that shipped four substyle profiles, ten canonical failure modes, four computable craft formulas, a 35-hit credential-validated corpus, a six-dimension critic loop, and Sacred Accident #20.

Published 2026-05-21. License: CC BY 4.0. Second WAR Room in the series; sibling of the Rap Excellence WAR Room.

Sacred Accident #20

Surfaced at §4 closing argument

“Authenticity is INHABITED, not INHERITED. A song is country when the narrator inhabits a coherent material world — not when the lyric collects country signals (truck + beer + dirt road + mama).”

The test: could the narrator be subbed into a city apartment without breaking ≥80% of the lyric? If yes, the song failed. If you swap the narrator out and the lyric breaks, the song earned its country identity. Full SA#20 entry →

The four country substyles

Country is not one style; it’s four commercially- distinct lanes. Per the Chartmetric 20K-song study + Hit Songs Deconstructed analyses of post-2013 #1s, each substyle has its own production envelope, lyric density profile, and failure modes. SongForgeAI routes prompts to one of the four via the C5 keyword + artist + structural detector.

Neo-traditional

75-95 BPM

CID ≥ 0.30 verses, bridge required, single narrator, narrative density

Reference artists: Zach Top · Tyler Childers · Cody Johnson · Sturgill Simpson

Radio-Country Crossover

95-120 BPM

Title repetition 6-9×, first chorus ≤ 0:50, post-chorus common

Reference artists: Morgan Wallen · Luke Combs · Kane Brown · Sam Hunt

Heartland Ballad

65-85 BPM

CID ≥ 0.35 verses (strictest), bridge load-bearing, LRR ≤ 0.45

Reference artists: Chris Stapleton · Lori McKenna · Brandy Clark · Mary Chapin Carpenter

Country Bar Anthem

110-140 BPM

Title repetition 8-10×, chant-friendly chorus, bridge often instrumental

Reference artists: Shaboozey · Dan + Shay · Garth Brooks · Old Crow Medicine Show

The ten canonical failure modes

The C3 Country Forbidden Archive ratifies the ten country-specific failure modes the WAR Room reduced from ~30 candidates. Each is detectable, genre-specific, high-frequency in generic-LLM country output, and pedagogically uncontroversial.

  1. #1
    Demographic Cosplay — Fake dialect used as country-genre flag rather than the narrator's actual tongue.
  2. #2
    Object List Verse — Pickup + beer + tailgate + bonfire without a narrator inhabiting them.
  3. #3
    Could-Be-Any-City Verse — Generic small town with no local detail.
  4. #4
    Stagnant Second Verse — V2 restates V1 with synonym swaps (VIG < 0.10).
  5. #5
    Title That Doesn't Carry — Title fails the bar test — stranger can't predict the song from the title.
  6. #6
    Bridge That Summarizes — Bridge restates chorus instead of pivoting.
  7. #7
    Telling Chorus — Names emotion ("I'm broken") instead of grounding it in sensory detail.
  8. #8
    Sentimentality Trap — Heartland-style writing reaching for mama/Bible/tears without earned specificity.
  9. #9
    Drift Narrator — POV shifts mid-song (he/I/we) without structural reason.
  10. #10
    Production-Crutch Hook — Chorus only works because of the production hook supporting it.

Full archive at docs/COUNTRY-FORBIDDEN-ARCHIVE.md. Each entry includes detection logic + fix patterns + substyle-specific severity.

Five computable craft formulas

The audit substrate is five primitives. Four are pure-sync (no model call, deterministic, fast). One is Haiku-judged (semantic recontextualization detection requires reading comprehension).

AbbrNameType
CID
Concrete Imagery Density
(concrete + sensory tokens) / total. Per-section thresholds per substyle.
pure-sync
LRR
Linguistic Repetitiveness Ratio
1 − (unique / total). Substyle classifier sanity check.
pure-sync
MSC
Metrical Symmetry Coefficient
V1 vs V2 line-by-line syllable parity. Target ≥ 0.90.
pure-sync
VIG
Verse Information Gain
New concrete tokens in V2 not in V1 or chorus / V2 total. Target ≥ 0.15.
pure-sync
SII
Semantic Inversion Index
Does the chorus mean something different after V2 than after V1? Haiku-judged.
Haiku-judged

Five-criteria success scorecard

Honest verdicts per WAR Room §5 criteria. Modified from the original Brett-graded plan (operator-side advisor relationship wasn’t available — substituted credential-validated corpus as calibration anchor).

§1

The Title Test

PASS

C12 title-first discipline gate makes the bar test system-enforced for all 4 country substyles. Forge prompt instructs the room to declare title FIRST + verify it predicts the song frame to a stranger.

§2

Substyle Coherence

PASS

Four audit primitives (CID/LRR/MSC/VIG) measure substyle adherence per-section. The C5 detector routes prompts to one of 4 substyles + the C10 plugin folds the primitives into one 0-100 country-craft score.

§3

Verified-Hit Calibration

PASS

35-hit credential-validated corpus (Billboard #1 / RIAA Platinum+ / Grammy / Diamond / canonical) at public/country-corpus-v1.json. C11.5 calibration ratchet enforces internal consistency. Top-credential songs MUST score A+ band (≥86) or the rubric has a bug — the "Brett Collapsed" failure SA#19 named.

§4

Verse Utility Test

PASS

C7 VIG primitive (Verse Information Gain) measures whether V2 introduces new concrete tokens not in V1/chorus. Threshold VIG ≥ 0.15. Pure-sync; runs on every country forge.

§5

The Authenticity Test

PARTIAL

The "sub the narrator into a city apartment" test is the C9 critic's culturalCoherence dimension (Haiku-judged). Without an external Nashville-pro advisor (operator constraint), full PASS requires sustained validation — currently calibrated against the 35-hit corpus. Future operator advisor relationship can flip this to full PASS.

The credential-validated calibration anchor

public/country-corpus-v1.json ships 35 verified country hits with structural metadata + craft analysis + expected rubric band. Each entry is credential-validated (Billboard #1 / RIAA Platinum+ / Grammy / Diamond / canonical-historical) so calibration doesn’t require a single-advisor relationship.

Distribution: neo-traditional 7, radio-country crossover 14, heartland ballad 10, bar anthem 4. The rule: any verified Billboard #1 + RIAA Platinum+ + Grammy song MUST score in the A+ band (≥86) on the rubric. If it scores below, that’s a rubric calibration bug — the SA#19 “Brett Collapsed” failure made measurable.

The C11.5 calibration ratchet (npm run check:country-calibration) enforces this contract in CI.

The 15-build arc

Mirror of the 20-build Rap Excellence WAR Room. Five builds faster because the substrate now exists (phonology library, genre-craft plugin registry, audit orchestrator pattern, public report convention).

C1 (B2896): Country Eval Context
C2 (B2897): Country Phonology Layer
C3 (B2898): Country Forbidden Archive (10 modes)
C4 (B2899): Country Substyle Registry (4 profiles)
C5 (B2900): Country Substyle Detector (3-tier router)
C6 (B2901): CID + LRR audit primitives
C7 (B2902): MSC + VIG audit primitives
C8 (B2903): SII Haiku-judged primitive
C8.5 (B2904): 35-hit Credential-Validated Corpus
C9 (B2905): Country Critic Loop (6 dimensions)
C10 (B2906): Genre Craft Plugin Registration
C11 (B2907): MODE_COUNTRY × 4 substyles in genre-modes
C11.5(B2908): Calibration Validator (ratchet 42)
C12 (B2909): Country Forge Amendments (title-first gate)
C13 (B2910): This page + SA#20 ratified

Methodology

The 100-expert × 100-round simulation drew on:

  • Nashville hit-writer consensus (Ashley Gorley, Hillary Lindsey, Shane McAnally, Lori McKenna, Brandy Clark, Luke Laird school)
  • Country songwriting pedagogy (Country Music Hall of Fame Words & Music curriculum, Pat Pattison object writing, Berklee prosody framework)
  • Quantitative analyses (Chartmetric 20K-country-song study, Hit Songs Deconstructed post-2013 deconstructions, 30-year Billboard year-end leaders)
  • Country authenticity scholarship (Evan Malone, Aaron Fox, Bill C. Malone, Diane Pecknold, Nadine Hubbs, Charles Hughes)
  • Industry gatekeeper criteria (BMG/Curb publisher evaluation, NSAI lyrical assessment)
  • The supplied operator research bundles (two LLM-prepared deep research artifacts informing Panel B/C/E synthesis)

The 100 rounds compressed to 25 contested questions; the consensus answers shipped as the 15-build sequence documented above.

Companion to the Rap Excellence WAR Room

The Rap Excellence WAR Room (B2838-B2865) ratified SA#19: a genre we cannot evaluate cannot be a genre we can serve. The Country deep-dive replicated the SA#19 template, proving the pattern repeats. SA#20 names the country-specific application: authenticity is inhabited, not inherited. Each subsequent genre deep-dive (Pop, Latin, R&B, Worship) will surface its own SA following the same template.

What this changes for country-writing users

Country forges now route to one of 4 substyle modes, generate against substyle-specific quantitative targets, get critiqued by the 6-dimension country critic, and score against the 35-hit credential-validated corpus. The title-first discipline gate is active. The bar test is system-enforced.

Published 2026-05-21 · Version 1.0 · CC BY 4.0 · Sibling of /reports/rap-excellence-war-room · SA#20 ratified.