Holographic dual to baryon Baryons can be effectively described at low energies as solitions in Skyrme model, that describe pions as NG bosons.
In Skyrme model exist current, that can be identified with baryon number current, which allow to identify solitons in Skyrme model with barions (David Tong: Lectures on Gauge Theory, section 5.3):
$$
B^\mu = \frac{1}{24\pi^2}\epsilon^{\mu\nu\rho\sigma} tr\left(U^\dagger \partial_\nu U U^\dagger \partial_\rho U U^\dagger \partial_\sigma U\right)
$$
So it leads to few questions about holographic description of this objects:
1) What is holographic dual to baryon?
2) What is holographic dual to pions?
3) What is holographic dual to Skyrme current?
 A: A proposal for the holographic dual of QCD is given by the so-called Witten-Sakai-Sugimoto model $[1,2,3]$.
The Witten background is a solution of a consistent truncation of Type IIA supergravity (where only the metric, the dilaton and $F_{(4)}$ are switched on) and it corresponds to $N_c$ D4-branes wrapped around $S^1$. Intuitively this $S^1$ gives an IR scale that should reproduce $\Lambda_{\mathrm{QCD}}$.
This was proposed by Witten as a model to study confinement in gauge theories but has some shortcomings (which also remain in the full Witten-Sakai-Sugimoto model). Those are 


*

*The presence of many Kaluza-Klein modes that transform under $\mathrm{SO}(5)$

*The absence of asymptotic freedom in the UV


So this model has to be taken as an effective description only for the low energy regime.
The background for just Yang Mills is
\begin{equation}
\mathrm{d} s^2 = \left(\frac{U}{R}\right)^{3/2}\bigl(\eta_{\mu\nu}\mathrm{d} x^\mu \mathrm{d} x^\nu + f(U) \mathrm{d}\tau^2\bigr) + \left(\frac{R}{U}\right)^{3/2}\left(\frac{\mathrm{d} U^2}{f(U)} + U^2 \mathrm{d} {\Omega_4}^2\right)\;,
\end{equation}
\begin{equation}
e^\phi =g_s \left(\frac{U}{R}\right)^{3/4}\;,\quad F_{(4)} = dC_{(3)} = \frac{2\pi N_c}{\mathrm{Vol}(S^4)}\,\omega_{S^4}\;,\quad f(U) = 1-\frac{U_\mathrm{KK}^3}{U^3}\;.\;,
\end{equation}
With the AdS/CFT dictionary
\begin{equation}
R^3 = \frac{9}{4}\;,\quad U_{\mathrm{KK}} = M_{\mathrm{KK}} =1\;,\quad g_s = \frac{1}{2\pi}\frac{g_{\mathrm{YM}}^2}{M_{\mathrm{KK}}l_s}\;, \quad\frac{2}{9} M_\mathrm{KK}^2 l_s^2 = \lambda^{-1}\;. \label{adscftdict2}
\end{equation}
Then in order to obtain QCD we need the quarks. This is done by introducing a stack of $N_f$ D8-branes and $N_f$ $\overline{\mathrm{D8}}$-branes that intersect the D4's. The strings with one end in the D8 and one in the D4 each transform as fundamental fermions.
More precisely, the D4 extend along the Minkowski four dimensions and on $\tau$ (compactified on $S^1$), while the D8 extend on Minkowski, the $\mathrm{AdS}_7$ radius $U$ and the four sphere.
From this one can introduce a probe D8 brane solution in the above background and study its DBI action
$$
S_{\mathrm{D8}} \propto \int \mathrm{d}^4x\, \mathrm{d}\tau \,\mathrm{d}\Omega_4\,\,e^{-\phi}\sqrt{\det g}\,.
$$
I am skipping many steps. But essentially, after dimensionally reducing one gets a five dimensional theory with some gauge fields. Then one can expand the gauge fields in some eigenfunctions that depend on the fifth coordinate. The coefficients are four dimensional fields that represent the mesons!
Furthermore, the super cool feature of this model is that the stable solution of the embedding of the D8 wants that the D8 and the $\overline{\mathrm{D8}}$ meet at a point $U_0$. Each of the brane stacks gives in the boundary a flavor group $\mathrm{U}(N_f)$, but when they join you only get the diagonal part. This is the chiral symmetry breaking effect!
Finally: baryons! They come from an idea of Witten $[4]$ of seeing them as branes wrapped around a compact manifold. In the case of the Sakai-Sugimoto model they can be interpreted as instantonic solutions of this five dimensional gauge field $[5,6]$. These solutions can be then quantized and the moduli space can be studied.
At the boundary these instantonic solutions become precisely the Skyrmion. The gauge symmetry in the bulk becomes a flavor global symmetry and the instantonic topological charge becomes the Skyrme topological charge.
So far the fermions are massless (because the D8 and D4 have no separation in $x^\mu$). But it's possible to introduce masses as well $[7]$.

$[1]\;\;$ E. Witten, “Anti-de Sitter Space, Thermal Phase Transition, And Confinement In
Gauge Theories,” Adv.Theor.Math.Phys 1998 505–532, hep-th/9803131v2.
$[2]\;\;$ T. Sakai and S. Sugimoto, “Low Energy Hadron Physics in Holographic QCD,” Prog.
Theor. Phys. 113 843–882, hep-th/0412141v5.
$[3]\;\;$ T. Sakai and S. Sugimoto, “More on a Holographic Dual of QCD,” Prog. Theor. Phys.
114 1083–1118, hep-th/0507073v4.
$[4]\;\;$ E. Witten, “Baryons And Branes In Anti de Sitter Space,” JHEP 1998 006,
hep-th/9805112.
$[5]\;\;$ H. Hata, T. Sakai, S. Sugimoto, and S. Yamato, “Baryons from instantons in
holographic QCD,” Prog. Theor. Phys. 117 1157–1180, hep-th/0701280v3.
$[6]\;\;$ K. Hashimoto, T. Sakai, and S. Sugimoto, “Holographic Baryons : Static Properties and
Form Factors from Gauge/String Duality,” Prog. Theor. Phys. 120 0806.3122v4
$[7]\;\;$ O. Aharony and D. Kutasov, “Holographic Duals of Long Open Strings,” Phys. Rev. D 78 026005, 0803.3547v2.
A: The user MannyC has given an excellent answer addressing the baryon construction. I would like to add some comments mostly to spark some
discussion and for completeness. Hopefully, the comments are interesting. 
In the following, I will also skip many steps, but if you are interested, please feel free to ask or comment. 
Since we are discussing fundamental degrees of freedom in the context of the AdS/CFT, I would like to refer to the paper of Karch/Katz (1) where the analyze the probe-brane construction that is explained in the accepted answer.  
After that paper, it was realized that the open-string fluctuations of the probe-brane would give matter in the adjoint representation of the $\mathcal{N}=2$ SYM, only now this adjoint matter would be comprised out of fields transforming in the fundamental representation of the gauge group. The full-fledged analysis of computing the mass spectra as a supergravity exercise can be found in the original paper by Myers et al (2). 
So far, we have a nice, clean example of adding fields transforming in the fundamental of the gauge group. We also understand that the addition of the probe-branes will break half of the supercharges if they are embedded appropriately/in the simplest way. We also know how to describe the dynamics of mesonic degrees of freedom and compute their mass spectra. 
The baryonic construction in the formal way has been described already and even in IIB the idea remains the same, of course following Witten's prescription.  
People wanted to understand whether they could describe chiral symmetry breaking in this flavoured version of the AdS/CFT or not and how that would be possible. It turned out that using a construction that was previously obtained by Constable and Myers (3) in IIB theory, Evans et al (4) studied the quark condensate and meson spectra of non-supersymmetric gauge theories holographically. In this latter paper, there is a discussion about the pion and the holographic interpretation -you can find the discussion on page 24. 
Let me go back to the baryons. Recently a paper appeared by Erdmenger et al (5). The original motivation was Beyond the Standard Model physics and generating composite fermionic states that can get light. 
However, the following was also found. In the simple D3-probe D7 brane system we have $\mathcal{N}=2$ supersymmetry. The bosonic mesons of the systems, due to SUSY, have fermionic partners -this is an effect due to SUSY and there is no analogue of fermionic mesons in QCD. The  spectra of these fermionic mesons (dubbed mesinos) can be found by means of representation theory and this is the analysis done in the paper by Myers that I have included. But, by doing so you cannot have access to their dynamics of course, which were needed in the final paper that I have included. The authors there observed that one family of these fermionic operators (the ones they call $\mathcal{G}$) are similar to the QCD baryon multiplet as the lowest entry in the multiplet is a state comprised of three elementary fermionic fields. They do point out, however, that this is not the true large-N baryon description. 
