\section{Cluster monomials for shifted quantum affine algebras}\label{sh:section} Throughout this section, $\g$ has finite type $A$, $D$, or $E$, and $q\in\C^*$ is not a root of unity. Let $I$ be the set of vertices of its Dynkin diagram, $C=(c_{ij})$ its Cartan matrix, $P$ its weight lattice, $\omega_i$ its fundamental weights, and $\alpha_i=\sum_jc_{ji}\omega_j$ its simple roots. Write $j\sim i$ when $c_{ij}=-1$, and let $W$ and $w_0$ be its Weyl group and longest element. Multiplication of shifted representation classes means fusion multiplication. \subsection{Highest weights, normalization, and formal series} For $i\in I$ and $a\in\C^*$, let $\Psi_{i,a}$ be the rational $\ell$-weight whose $i$th component is $1-au$ and whose other components are $1$. A \emph{pure divisor} is a finite Laurent monomial in these $\Psi_{i,a}$. For $\lambda\in P$, the symbol $[\lambda]$ denotes the constant $\ell$-weight with components $q^{\langle\lambda,\alpha_i^\vee\rangle}$, and also the corresponding invertible one-dimensional class. Set $\overline P=\frac12P$ and extend scalars to $\Z[\overline P]$ when normalizing cluster variables. This is the coefficient extension in \cite[\S7.1 and Theorem~9.15]{GHL24}. Every rational highest $\ell$-weight $h$, regular and nonzero at $u=0$, defines a simple shifted module $L(h)$. Its shift is determined by the degrees of the components of $h$. The highest $\ell$-weight space has dimension one, and all its other $\ell$-weights are obtained by multiplication by inverse root monomials \begin{equation}\label{sh:root} Y_{i,a}=[\omega_i]\frac{\Psi_{i,aq^{-1}}}{\Psi_{i,aq}},\qquad A_{i,a}=Y_{i,aq^{-1}}Y_{i,aq}\prod_{j\sim i}Y_{j,a}^{-1}. \end{equation} These facts, the fusion multiplication, and the injectivity of the $q$-character map are established in \cite{H23}; see also \cite[\S9.1]{GHL24}. For any rational $\ell$-weight $h$, the symbol $[h]$ denotes its basis monomial in the $\ell$-weight group algebra. We use \begin{equation}\label{sh:normalized-character} F(h)=\frac{\chq L(h)}{[h]}. \end{equation} Constant weight twists leave $F(h)$ unchanged. Choose a bipartite coloring $\epsilon_i\in\{0,1\}$ and put \[ V=\{(i,r)\in I\times\Z:r\equiv\epsilon_i\pmod2\},\qquad P_{i,r}=\Psi_{i,q^r}. \] Let $\cP_V$ be the free abelian group on the $P_{i,r}$, $(i,r)\in V$. For $p=\prod_{i,r}P_{i,r}^{e_{i,r}}\in\cP_V$, define \begin{equation}\label{sh:omega} \Omega(p)=\sum_{i,r}\frac r2e_{i,r}\omega_i, \qquad \nu(p)=[-\Omega(p)]p, \qquad \Supp_\Psi(p)=\{r:e_{i,r}\ne0\text{ for some }i\}. \end{equation} We set $\max\Supp_\Psi(1)=-\infty$. The normalization $\nu$ is multiplicative. In particular, if \begin{equation}\label{sh:basic-divisor} d_{i,r}=\frac{P_{i,r+2}}{P_{i,r-2}} \prod_{j\sim i}\frac{P_{j,r-1}}{P_{j,r+1}}, \end{equation} then $\Omega(d_{i,r})=\alpha_i$ and $\nu(d_{i,r})=A_{i,q^r}^{-1}$. \begin{lemma}\label{sh:independence} The root monomials $A_{i,a}$ are multiplicatively independent for finite-support integer exponents. Consequently their nonnegative monoid is free and pointed. \end{lemma} \begin{proof} Partition $\C^*$ into its $q^\Z$-orbits. Each root monomial involves only one orbit, so it suffices to work on an orbit $a_0q^\Z$. Encode a finite exponent sequence by a Laurent polynomial in $z$, with multiplication by $z$ representing an increase of the spectral exponent by one. The map from the exponents of the inverse root monomials to their pure $\Psi$-divisors is the matrix \[ B(z)_{ii}=z^2-z^{-2},\qquad B(z)_{ij}=z^{-1}-z\quad(j\sim i),\qquad B(z)_{ij}=0\quad(i\ne j,\ j\not\sim i). \] At $z=1$ its expansion is $B(z)=2(z-1)C+O((z-1)^2)$. Since $\det C\ne0$, $\det B(z)$ is a nonzero Laurent polynomial. The matrix is therefore injective on $\Z[z,z^{-1}]^I$: multiply a vector in its kernel by its adjugate and use that $\Z[z,z^{-1}]$ is an integral domain. A multiplicative relation between full root monomials would induce a relation between their pure divisors, and thus has all exponents zero. A product of nonnegative powers can equal $1$ only when every power is zero, which proves pointedness. \end{proof} Put $t_{i,a}=A_{i,a}^{-1}$. Let $\mathscr M$ be the free commutative monoid of finite-support nonnegative exponent vectors in the variables $t_{i,a}$, and use the ring \begin{equation}\label{sh:completion} \mathscr R=\prod_{m\in\mathscr M}\Z m,\qquad (fg)_m=\sum_{m_1m_2=m}f_{m_1}g_{m_2}. \end{equation} Each sum in this definition is finite. This ring allows infinitely many different spectral variables in one total degree. A series with constant coefficient $1$ is invertible: in $(1+h)^{-1}=\sum_{k\ge0}(-h)^k$, the coefficient of a monomial of total degree $d$ receives contributions only from $k\le d$. For a finite colored spectral set $S\subset I\times\C^*$, let \[ \pi_S:\mathscr R\longrightarrow\Z[[t_{i,a}:(i,a)\in S]] \] set all variables outside $S$ equal to zero. This is a ring homomorphism and commutes with inversion of constant-one series. The maps $\pi_S$ jointly separate elements of $\mathscr R$, since every monomial has finite support. Formula \eqref{sh:normalized-character} belongs to $\mathscr R$, has constant coefficient $1$, and has nonnegative integer coefficients. \subsection{Invariance under local units} We will use the local description of shifted simple modules in \cite[Corollary~3.9 and equations~(132)--(136)]{HN26}. We specify the coordinate convention. For an $\ell$-weight $h(u)$, the corresponding function in the coordinate of that paper is $\psi(z)=h(z^{-1})$. Thus $1-au$ becomes $1-a/z$, and its zero is the local point $z=a$. The root parameter $a$ in $A_{i,a}$ is unchanged. \begin{lemma}[Local units]\label{sh:local-unit} Let $h$ and $g$ be rational $\ell$-weights. Suppose that, for every $(i,a)\in S$, the function $g_i(z^{-1})$ is regular and nonzero at $z=a$. Then \begin{equation}\label{sh:local-unit-equality} \pi_S F(hg)=\pi_S F(h). \end{equation} In particular, if every $h_i(z^{-1})$ is regular and nonzero at the points of $S$ of color $i$, then $\pi_SF(h)=1$. \end{lemma} \begin{proof} Fix a colored multiset $\mathbf x=((i_1,x_1),\ldots,(i_n,x_n))$ supported in $S$, with $n_i$ entries of color $i$. Repetitions of a spectral point are allowed. Let $\mathcal S_{-\mathbf n}$ denote the corresponding negative shuffle space. It is the fixed space, independent of $h$, in the shuffle realization used in \cite[\S2]{HN26}. Its elements are rational functions, symmetric in the variables of each color, with products of linear differences as denominators. The local quotient theorem identifies the multiplicity of $\prod_{b=1}^nt_{i_b,x_b}$ in $F(h)$ with \[ \dim_\C\bigl(\mathcal S_{-\mathbf n}/J(h)_{\mathbf x}\bigr). \] Here $J(h)_{\mathbf x}$ consists of the shuffle functions $f$ for which \begin{equation}\label{sh:local-residue} \Res_{z_n=x_n}\cdots\Res_{z_1=x_1} \frac{f(z_1,\ldots,z_n)\,p(z_1,\ldots,z_n) \prod_{b=1}^n h_{i_b}(z_b^{-1})} {\prod_{b0$. \subsection{Finite sweeps and the entire upper region} We prove a uniform comparison with an actual finite mutation seed. All vertices in the following statement retain their physical labels in $V$. An ideal inserted seed is the seed defined above; an actual seed is obtained from $\Gamma_e$ by finitely many ordinary cluster mutations. Only the latter uses \eqref{sh:finite-input}. \begin{proposition}[Finite patch]\label{sh:patch} Fix a reduced word $\mathbf i=(i_1,\ldots,i_\ell)$ and separated insertion heights $r_1,\ldots,r_\ell$. Let $(a^{(m)},p^{(m)})$ be the ideal seed after its first $m$ insertions. There are constants $B_m,C_m$, independent of the cutoffs below, with the following property. Choose \[ R_m\equiv r_m\pmod2,\qquad R_1\ll R_2\ll\cdots\ll R_\ell, \] all sufficiently negative. At stage $m$, set $i=i_m$ and $n=n_i^{(m-1)}$, and, in the actual seed from the preceding stage, mutate successively at \begin{equation}\label{sh:sweep} (i,R_m-2n),\ (i,R_m-2n+2),\ \ldots,\ (i,r_m-2n-2). \end{equation} This is a finite sequence of $(r_m-R_m)/2$ mutations. The resulting actual seed $(a^{\rm act},p^{\rm act})$ satisfies, with $U_m=\{(j,r)\in V:r\ge R_m+C_m\}$: \begin{enumerate}[label=\textup{(\roman*)},leftmargin=2.7em] \item\label{sh:patch-I} $a^{\rm act}_{uv}=a^{(m)}_{uv}$ for every pair $u,v\in U_m$; \item\label{sh:patch-II} $\prod_up_u^{{\rm act}\,a^{\rm act}_{vu}} =\prod_u(p_u^{(m)})^{a^{(m)}_{vu}}$ for every $v\in U_m$; \item\label{sh:patch-III} $\Supp_\Psi(p_v^{\rm act}/p_v^{(m)})\subseteq(-\infty,R_m+C_m]$ for every $v\in U_m$; \item\label{sh:patch-IV} for every $(j,r)\in V$, \begin{equation}\label{sh:global-support} \max\Supp_\Psi(p_{j,r}^{\rm act}) \le\max\{r,R_m\}+B_m. \end{equation} \end{enumerate} The products in \textup{(ii)} include all neighbors, including those outside $U_m$. \end{proposition} \begin{proof} We first establish bounds for the ideal seeds, including the intermediate seeds in which the newest insertion moves upward. At most $\ell$ insertions occur. Each insertion changes a physical coordinate by either zero or two and introduces only the pattern in the insertion rule. Hence there is a constant $L_q$ such that \begin{equation}\label{sh:ideal-geometry} a^{\rm ideal}_{(j,r),(k,s)}\ne0\quad\Longrightarrow\quad |r-s|\le L_q \end{equation} for every intermediate ideal seed. For example, increasing a bound by four at each insertion bounds the change of every old arrow length and the lengths of the newly introduced arrows. The bound does not depend on the location of the insertion. A label is obtained from a $P_{j,s}$ by at most $\ell$ operators \eqref{sh:braid}, each of which changes a spectral index by at most two, and by at most $\ell$ coordinate changes of size two. Consequently there is a constant $L_p$ such that \begin{equation}\label{sh:ideal-support} \max\Supp_\Psi(p_{j,r}^{\rm ideal})\le r+L_p \end{equation} for all these seeds. Choose the fixed gaps between the $r_m$ large enough that the local pattern of each moving insertion, throughout its upward motion to $r_m$, stays below all preceding insertion patterns. The bounds just obtained allow one choice of gaps for the entire word. Set $B_0=C_0=0$. For $m=0$, actual and ideal seeds are both $\Gamma_e$, with $p_{j,r}=P_{j,r}$. They have identical adjacency and full divisors, all label ratios are $1$, and the global support bound is exactly $r$. Thus all four assertions begin with exact equalities. Suppose that they have been proved after stage $m-1$, and put $R=R_m$, $i=i_m$, and $w=w^{(m-1)}$. Choose $R$ above the preceding boundary by a fixed amount so that \begin{equation}\label{sh:cutoff-gap} R-2\ell\ge R_{m-1}+C_{m-1}. \end{equation} For $m=1$ this requirement is empty. Use the prefix heights $h(j,r)=r+2n_j^{(m-1)}$. Let $c_0$ exceed the finitely many offsets in the insertion pattern and the bounds needed to exclude its other-color endpoints, and set \begin{equation}\label{sh:sector} S_R=\{(i,r):h(i,r)\ge R\} \ \cup\!\!\bigcup_{j\ne i}\{(j,r):h(j,r)\ge R+c_0\}. \end{equation} Every vertex mutated in \eqref{sh:sweep} belongs to $S_R$. The prefix ideal seed and the ideal seed with a new insertion initially at height $R$ have identical labels and principal matrices on $S_R$. The new green vertex and the shifted lower vertices are outside $S_R$, as are the changed other-color endpoints. By \eqref{sh:cutoff-gap}, the entire set $S_R$ belongs to the preceding upper region. Actual and moving ideal principal matrices therefore agree on $S_R$ before the sweep. \emph{Principal matrices.} For $u,v,k\in S_R$, the mutated entry $a'_{uv}$ depends only on $a_{uv},a_{uk},a_{kv}$. Restricting to $S_R$ consequently commutes with mutation at $k\in S_R$. Induction over the finite sequence \eqref{sh:sweep} proves equality of the two principal matrices on \emph{all} of $S_R$ at every step. The moving ideal seed ends at the ideal insertion of height $r_m$, by the red-vertex mutation rule. This proves assertion~(i) on any upper region contained in $S_R$. \emph{Full front divisors and dominant branches.} The current-color vertices at heights $R,R+2,\ldots,r_m$ form an upward path in the principal quiver on $S_R$. Only this color is mutated in the stage. At its first vertex all remaining path arrows point toward higher vertices. Upon mutating that vertex the one edge to the next vertex reverses; there is no directed two-path through it between two unmutated path vertices. Inductively, just before mutating the next vertex, its edges to the already mutated part point backward and its edge to the remaining part points forward. No arrow from a future front enters the current front. Equivalently, if $k$ is the current front and $j$ a future front, $a_{kj}\ge0$. This path calculation remains valid in the full quiver, because principal restriction to the current-color set commutes with its mutations. Initially, all actual front divisors equal the corresponding prefix ideal divisors by assertion~(ii) of stage $m-1$ and \eqref{sh:cutoff-gap}. Each is $T'_wd_{i,a}$ at its height $a$. At the first ideal red front the same equality follows from \eqref{sh:red-D}. Apply Lemma~\ref{sh:dominance} to the first actual exchange. Formula \eqref{sh:front-root} selects the incoming branch. Formula \eqref{sh:D-incoming}, together with $a_{kj}\ge0$, then leaves every future-front divisor unchanged. Thus at the next front the actual full divisor is still its original prefix divisor, and the same argument again selects the incoming branch. Induction gives, at every front of height $a$, \begin{equation}\label{sh:all-fronts} D_k^{\rm act}=D_k^{\rm ideal}=T'_wd_{i,a}, \qquad \nu(D_k^{\rm act})\in\mathscr M\setminus\{1\}. \end{equation} The ideal label update is \eqref{sh:red-update}, so its incoming updates are exactly the specified chamber labels. This proves \eqref{sh:all-fronts} using the full divisors, including any actual neighbors below $S_R$. \emph{Full divisors above a fixed strip.} Before the sweep, actual and prefix ideal divisors agree throughout $S_R$. Comparing the prefix ideal seed with the newly inserted ideal seed changes full divisors only within a bounded strip of height $R$. To see this, above the strip the labels agree, and \eqref{sh:ideal-geometry} places every ideal neighbor above the portion whose labels or arrows changed. Therefore $D_j^{\rm act}/D_j^{\rm ideal}=1$ for every $j\in S_R$ above that strip. At a front the ratio is also one by \eqref{sh:all-fronts}. When a front $k$ is mutated, the ratio at any other $j\in S_R$ is unchanged: the exponents in \eqref{sh:D-incoming} agree by principal-matrix equality, and the multiplier $D_k$ agrees by \eqref{sh:all-fronts}. At $k$ both divisors invert. Thus all full divisors initially equal above the strip remain equal there throughout the sweep. One fixed upper threshold proves assertion~(ii), uniformly over all vertices above it. \emph{Global support.} We next prove a bound independent of the number of fronts. For a current front $k=(i,r)$, every neighbor outside $S_R$ has physical coordinate at most $R+c$ for a fixed constant $c$ depending only on $c_0$ and $\ell$. Such a neighbor is never mutated during this stage. By the preceding global bound, its entire label is supported below $R+c+B_{m-1}$; the inequality $R>R_{m-1}$ is used here. An incoming neighbor of another color inside $S_R$ is also unchanged. Principal-matrix equality and \eqref{sh:ideal-geometry} put its coordinate at most $r+L_q$, so its support is at most $\max\{r,R\}+L_q+B_{m-1}$. An already mutated same-color incoming neighbor, if present, has smaller physical coordinate than $r$. Its inductive bound is therefore at most $\max\{r,R\}+B_m$. The path argument excludes incoming same-color neighbors still to be mutated. Finally, the old denominator label $p_k$ is unmutated and satisfies the previous bound. Choose $B_m\ge B_{m-1}+\max\{c,L_q,0\}$, increasing it by the fixed coordinate offsets if necessary. The incoming update \eqref{sh:incoming-update} multiplies and divides Laurent monomials. The support of such a product is contained in the union of the supports of its factors; no new spectral position is introduced. The bound $\max\{r,R\}+B_m$ consequently holds at this front. Induction over the fronts uses the same $B_m$ at each step. Unmutated vertices retain the previous bound, which is at most \eqref{sh:global-support} because $R_m>R_{m-1}$ and $B_m\ge B_{m-1}$. This proves assertion~(iv) on the whole quiver. \emph{Ratios of labels.} Initially, all actual-to-ideal ratios on $S_R$ are supported below $R_{m-1}+C_{m-1}\le R$. At a front the two incoming updates have identical exponents for neighbors in $S_R$. Their quotient is a product of previously existing internal ratios, divided by the old front ratio, and of factors from neighbors outside $S_R$. The latter actual factors are supported below $R+c+B_{m-1}$, and the ideal factors are supported below $R+c+L_p$, by \eqref{sh:ideal-support}. There is consequently a constant $E_m$, independent of the front and the sweep length, such that every one of these external factors is supported below $R+E_m$. Induction over the incoming updates preserves the same bound on all internal ratios, because multiplication and division only take unions of supports. Unmutated internal ratios already satisfy it. This proves assertion~(iii) on $S_R$ with threshold $R+E_m$. Choose $C_m$ large enough that $U_m\subseteq S_R$, that $R+C_m$ lies above the full-divisor exceptional strip, and that $C_m\ge E_m$. Assertions~(i)--(iv) now hold together after stage $m$. Their constants depend on the fixed word and ideal patterns and not on any cutoff or sweep length. This completes the induction. \end{proof} \begin{corollary}\label{sh:remote-boundary} In Proposition~\ref{sh:patch}, take even integers $N\to\infty$ and \begin{equation}\label{sh:cutoff-family} R_m(N)=-(\ell-m+1)N+\delta_m, \end{equation} where the bounded integers $\delta_m$ impose the required parities. Every fixed finite set of vertices is eventually contained in $U_\ell$. Any additional actual neighbor of such a set, absent from the ideal quiver, has its entire pure label supported below a bound tending to $-\infty$. \end{corollary} \begin{proof} All fixed gap inequalities in the proposition hold for sufficiently large $N$, and $R_\ell(N)+C_\ell\to-\infty$. If both endpoints of an arrow lie in $U_\ell$, its signed multiplicity agrees with the ideal one by assertion~(i). Thus an additional neighbor has physical coordinate $rw$ and an exponent $r$. Choose a reduced word for $w$ and separated insertions whose lower tail contains the local pattern of an additional color-$i$ insertion with red height $r-2$. The preceding insertion heights can be chosen sufficiently large for every specified $r$. Apply Proposition~\ref{sh:patch} to this seed and retain the one red exchange at that vertex. Fix a finite colored spectral set $S$ containing the root support of $T'_w(A_{i,q^{r-2}}^{-1})$. For sufficiently large $N$, the central labels have ratios to the ideal labels which are units at $S$, by assertion~(iii). Every additional actual neighbor has a whole label which is a unit at $S$, by Corollary~\ref{sh:remote-boundary}. After dividing the actual exchange by its dominant highest monomial, Lemma~\ref{sh:local-unit} identifies the normalized characters of the central factors with those of their ideal pure labels, and sends the normalized characters of additional factors to $1$. Assertion~(ii) supplies the exact coefficient \[ D_r=T'_w(A_{i,q^{r-2}}^{-1}). \] The highest-label update at the red vertex is \eqref{sh:red-update}. Thus, writing $F_{w,i,r}=F(T'_wP_{i,r})$, the resulting equality is the restriction under $\pi_S$ of \begin{equation}\label{sh:simple-QQ} F_{ws_i,i,r} =\frac{\prod_{j\sim i}F_{w,j,r-1}}{F_{w,i,r-2}} +D_r\frac{F_{w,i,r}}{F_{w,i,r-2}}F_{ws_i,i,r-2}. \end{equation} To justify the newly mutated label in this passage, its quotient with $T'_wT'_iP_{i,r}$ is the quotient of the two incoming highest updates. Its internal factors are the central label ratios, and its external factors are the additional neighbor labels. All are units at $S$. This proves the stated application of Lemma~\ref{sh:local-unit} to that label as well. Enlarging $S$ by any finite set and using separation of coefficients proves \eqref{sh:simple-QQ} in $\mathscr R$. The same normalized recurrence follows from \eqref{sh:QQ} for $\mathcal Q_{w,i,r}$: the two dominant pure highest products agree by \eqref{sh:red-update}, and the ratio of the other product is $D_r$ by \eqref{sh:red-D}. The actual front argument in the proof of Proposition~\ref{sh:patch} shows that $D_r$ is a nontrivial inverse-root monomial of total degree $d=\operatorname{ht}(w\alpha_i)>0$. We now prove uniqueness, which will also identify the prescribed series with the simple characters. Proceed by reduced length. For $w=e$, both families are $1$, since $L(P_{i,r})$ is one-dimensional. Suppose that all chamber series of length at most $\ell(w)$ have been identified. Appending $s_i$ changes only the chamber $w\omega_i$, since $s_i\omega_j=\omega_j$ for $j\ne i$. Subtract the two instances of \eqref{sh:simple-QQ} and set \[ \Delta_r=F_{ws_i,i,r}-\mathcal Q_{ws_i,i,r},\qquad V_r=F_{w,i,r}/F_{w,i,r-2}. \] The already identified terms cancel, leaving $\Delta_r=D_rV_r\Delta_{r-2}$. For every $K\ge1$, \begin{equation}\label{sh:QQ-uniqueness} \Delta_r= \left(\prod_{b=0}^{K-1}D_{r-2b}V_{r-2b}\right)\Delta_{r-2K}. \end{equation} Each $V_{r-2b}$ belongs to $\mathscr R$ and has constant term one. Every monomial on the right has total degree at least $Kd$. Given a fixed monomial of degree $d_0$, choose $K>d_0/d$; its coefficient in $\Delta_r$ is then zero. All coefficients vanish, proving \eqref{sh:Q-simple-equality}. \end{proof} The ADE simplicity of these $Q$-variables is also a consequence of \cite{KLLLPW26}; see \cite[Remark~11.15(iii)]{GHL26}. The preceding argument provides the initial-character identification needed for the finite-mutation proof below. \subsection{Finite mutation words and full cluster monomials} \begin{proposition}\label{sh:finite-word} Let $\Sigma$ be a separated inserted seed of $\cA_{w_0}$ and let $\boldsymbol\mu=(k_1,\ldots,k_t)$ be any finite mutation sequence. Every variable $X$ of $\boldsymbol\mu(\Sigma)$ satisfies $\chq\mathcal I(X)=\nu(p)F(p)$ for a pure divisor $p$. Moreover, every cluster monomial $M$ in that seed satisfies \begin{equation}\label{sh:monomial-character} \chq\mathcal I(M)=\nu(p_M)F(p_M), \qquad p_M=\prod_vp_v^{e_v} \quad\text{if }M=\prod_vX_v^{e_v}. \end{equation} \end{proposition} \begin{proof} Choose a finite set $U$ containing all mutated vertices, all of their ideal neighbors at each step of $\boldsymbol\mu$, and all vertices occurring in $M$. Such a set is finite: the quiver is locally finite, each mutation changes finitely many entries, and $M$ has finite support. Use the actual finite seeds of Proposition~\ref{sh:patch} with cutoffs \eqref{sh:cutoff-family}, and then apply the same word $\boldsymbol\mu$ to them. For sufficiently large $N$, all vertices of $U$ and their specified ideal neighbors lie in the entire upper region $U_\ell$. We give the simultaneous induction over this fixed word. Initially, the ideal variables have the stated simple-character values by Proposition~\ref{sh:Q-simple}. On the whole upper region the actual and ideal principal matrices agree, their full exchange divisors agree, and their pure highest-label ratios are supported below a bound tending to $-\infty$. No vertex outside $U_\ell$ is mutated. Its actual label is therefore unchanged during the word and is supported below the bound in Corollary~\ref{sh:remote-boundary}. Principal restriction commutes with all mutations in the word, so the actual and ideal matrices continue to agree on the whole upper region. In particular, every additional actual neighbor of a vertex of $U$ remains outside this region, even if that neighbor first becomes adjacent during the word. Suppose the assertions have been proved before the next mutation at $k$. Let $D_k$ be the common full pure exchange divisor. The actual positive exchange and Lemma~\ref{sh:dominance} imply that exactly one of $\nu(D_k)$ and $\nu(D_k)^{-1}$ is a nontrivial inverse-root monomial. Since $D_k$ is the fixed ideal divisor, this choice is independent of $N$. Use the corresponding incoming or outgoing update to define the new ideal pure label $p'_k$. The quotient of the actual and ideal updated labels is a product of old internal label ratios and unchanged external actual labels, divided by the old label ratio at $k$. All have support below one bound tending to $-\infty$. Only finitely many such updates are made, so this assertion persists throughout the word. Formulas \eqref{sh:D-incoming} and \eqref{sh:D-outgoing} propagate full-divisor equality at every vertex of $U$: their exponents agree by principal equality and their multiplier $D_k$ agrees by the induction hypothesis. The divisor at $k$ inverts on both sides. It remains to identify the character of the new ideal label with the prescribed mutated variable. For incoming dominance the normalized actual exchange is \begin{equation}\label{sh:normalized-exchange} F(p_k^{\rm act})F((p'_k)^{\rm act}) =\prod_v F(p_v^{\rm act})^{[-a^{\rm act}_{kv}]_+} +\nu(D_k)\prod_v F(p_v^{\rm act})^{[a^{\rm act}_{kv}]_+}. \end{equation} For outgoing dominance interchange the two products and replace $\nu(D_k)$ by $\nu(D_k)^{-1}$. Fix a finite colored spectral set $S$, enlarged to contain the support of this root coefficient. All central actual-to-ideal ratios and all external highest labels are units at $S$ for sufficiently large $N$. Lemma~\ref{sh:local-unit} turns \eqref{sh:normalized-exchange} into the same formula for the simple normalized characters $F(p_v)$ and $F(p'_k)$, with external factors equal to $1$. By the induction hypothesis, the old ideal variables have their prescribed $q$-character values. Cluster mutation gives exactly this normalized formula for the new prescribed image. Division by $F(p_k)$ is legitimate in \eqref{sh:completion}, and commutes with $\pi_S$. Thus the new prescribed $q$-character is $\nu(p'_k)F(p'_k)$ after restriction to every $S$, and hence without restriction. This closes the induction for all variables and all full divisors required by the word. For the monomial assertion, retain the actual monomial with the same exponents in the final actual cluster. Put \[ p_{M,N}=\prod_v(p_{v,N}^{\rm act})^{e_v},\qquad p_M=\prod_vp_v^{e_v}. \] By \eqref{sh:finite-input}, the actual monomial is a simple class. Its highest monomial is $\nu(p_{M,N})$, so multiplicativity of $q$-characters gives the equality of normalized series \begin{equation}\label{sh:whole-actual-product} F(p_{M,N})=\prod_vF(p_{v,N}^{\rm act})^{e_v}. \end{equation} The whole ratio $p_{M,N}/p_M$ is a finite product of the remote ratios from the induction, and is therefore a unit at each fixed $S$ for sufficiently large $N$. Apply Lemma~\ref{sh:local-unit} to this whole ratio on the left of \eqref{sh:whole-actual-product}, and to each factor ratio on the right. It gives \[ \pi_SF(p_M)=\pi_S\prod_vF(p_v)^{e_v}. \] Every root monomial is detected by some finite $S$, so $F(p_M)=\prod_vF(p_v)^{e_v}$ in $\mathscr R$. Multiplying by $\nu(p_M)=\prod_v\nu(p_v)^{e_v}$ proves \eqref{sh:monomial-character}. \end{proof} \begin{proof}[Proof of Theorem~\ref{thm:shifted-main}] Choose the reduced expression used by the original GHL initial seed, and change its insertion parameters to obtain a separated seed. Every cluster monomial of $\cA_{w_0}$ is reached from this seed by a finite mutation sequence: changing the insertion parameters uses finitely many of the normalized $QQ$ exchanges, and cluster mutation itself uses finite words. Proposition~\ref{sh:finite-word} and injectivity of the $q$-character map therefore give, for every cluster monomial $M$, \[ \mathcal I(M)=[-\Omega(p_M)]\,[L(p_M)]. \] Apply the same proposition to $M^2$, which is a monomial in the same cluster. Its pure highest weight is $p_M^2$, and hence \[ [L(p_M)]^2=[L(p_M^2)]. \] The fusion square has finite length by \cite{HZ25}. Its class is the class of a single simple module, so its length is one and $L(p_M)$ is real. The quiver and embedding in \cite[Conjecture~9.10]{HICM26} are the primed quiver $\Gamma'_\infty$ and the original embedding \eqref{sh:prescribed-embedding} for $\cA_{w_0}$. \end{proof}