PREVIOUS STATE
GHOST
P2 · observed September 19, 2019
Open source evidencePermanent Signal
chg-ghost-legend-v1-to-v4
GHOST Legend: Declared amount change
CONFIRMEDMAJOR materiality, recorded editoriallyReviewed July 29, 2026
Across the verified V1 and V4 states L-citrulline increased 4 g → 6 g; beta-alanine increased 2 g → 3.2 g; total caffeine increased 202 mg → 300 mg; the declared formula architecture changed materially.
FormulaSignal observed the earlier state on September 19, 2019 and the newer state on July 23, 2026. Those are observation dates, not the date the brand made the change.
Before and after
The material differences.
Declared facts and deterministic calculations remain labeled by their evidence.
- Previous state observed
- September 19, 2019
- New state observed
- July 23, 2026
- Signal type
- Declared amount change
- Record category
- FORMULA
DECLARED FACT COMPARISON
Beta-alanine: 2 g to 3.2 g (+60%).
DECLARED FACT COMPARISON
L-citrulline: 4 g to 6 g (+50%).
DECLARED FACT COMPARISON
Caffeine: 202 mg to 300 mg (+48.5%).
RECORDED COMMERCIAL CONTEXT
L-citrulline increased 50%; beta-alanine increased 60%; caffeine increased 48.5% across the endpoints.
Source evidence
Two dated endpoints bound the finding.
NEW STATE
GHOST
P1 · observed July 23, 2026
Open source evidenceLimits and review
What this Signal does not establish.
Limitations
- The evidence establishes an observation window, not an exact reformulation date.
- An observation date is the date FormulaSignal read a source. It is not a manufacture date and not the date a change was made.
- V2 and V3 existed between these endpoints. This is a lineage comparison, not one direct reformulation event. The endpoints are different flavors, Grapefruit in 2019 and Blue Raspberry in 2026. The compared actives are declared per serving in both, so the amounts are like for like, but flavor-specific excipients are outside this comparison.
- At least one endpoint did not record its claim set in full, so a claim present on one side and absent on the other is non-disclosure rather than removal.
- At least one endpoint did not record its certification set in full, so a certification present on one side and absent on the other is non-disclosure rather than removal.
- Review state
- CONFIRMED for public use
- Reviewed by
- FormulaSignal
- Review recorded
- July 29, 2026
- Evidence basis
- Two dated source endpoints
- Released
- July 29, 2026. Record V1 releases a Signal on the date FormulaSignal records human review. No separate publication timestamp is recorded.
Continue through the Record