From 844caf477de0ec27ca82ea86a92cdb35fb8de4f8 Mon Sep 17 00:00:00 2001
From: admin <peancharoen.pslcp3@gmail.com>
Date: Sun, 19 Apr 2026 11:09:35 +0700
Subject: [PATCH] 690419:1109 feat: update CI/CD to use SSH key authentication
 #02

---
 .../LCBP3_RAG_Implementation_Guide.docx       | Bin 20347 -> 0 bytes
 .../LCBP3-RAG-Implementation-Guide-v1.0.0.md  | 544 ++++++++++++++++++
 .../LCBP3-RAG-Implementation-Guide-v1.1.0.md} |   0
 .../LCBP3-RAG-Implementation-Guide-v1.1.1.md} |   0
 4 files changed, 544 insertions(+)
 delete mode 100644 specs/06-Decision-Records/ADR-022-/LCBP3_RAG_Implementation_Guide.docx
 create mode 100644 specs/06-Decision-Records/ADR-022-Retrieval-Augmented-Generation/LCBP3-RAG-Implementation-Guide-v1.0.0.md
 rename specs/06-Decision-Records/{ADR-022-/LCBP3 RAG Implementation Guide v1.1.0.md => ADR-022-Retrieval-Augmented-Generation/LCBP3-RAG-Implementation-Guide-v1.1.0.md} (100%)
 rename specs/06-Decision-Records/{ADR-022-/LCBP3 RAG Implementation Guide v1.1.1.md => ADR-022-Retrieval-Augmented-Generation/LCBP3-RAG-Implementation-Guide-v1.1.1.md} (100%)

diff --git a/specs/06-Decision-Records/ADR-022-/LCBP3_RAG_Implementation_Guide.docx b/specs/06-Decision-Records/ADR-022-/LCBP3_RAG_Implementation_Guide.docx
deleted file mode 100644
index 5f5da2198e91c6bc7d6a1f173c4c05be5e2e1480..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 20347
zcmd43WppG<(j}N;W@g3`Gcz+YGqY4uiJ6&IVrHfiGcz+&iK)b}dfnga)7`VXXa3D(
zobb$u^z`FtcHQi{yICkmgMdN<{(AYtCF=a+!~YauU+>@TolNKz{y(e0{9Q%g$<)UA
zUlk$$(U!(IoIL^%06_HB5bl3gG_f~!wKcVKp>wyjq5JEnRY{X_{S1gf`ee6R98?O>
z<?Ap)r)a?t4i}TLVZJ8zNZy`NbB%Q6H1K>=E7{MDS{|M6YQ{HUQwU<rN~vg60x%>H
zn~b-!9h+4nZVf)6@y4yYue~qCJhR#H__JLKr=kcEuX7qRyM?TBj<9pC3r;tq+_{nE
zT1pt^ac2`o6j+Yjnu+<GT2J!Lmg|B&6-W?Z*!C(4vBOAS5(m{FCMI*HBSiOZ@5JVv
ze5Jgt48t(Jt5Sj+$3Yg+g0>9!{a+V62SEVZPlgFO=v~G$I)G7{X9pYwMR4szIAPQa
zZv>(lMeubJN*)AVuP%UGx8z0rR}151j!XR~r+xQO$dz~)lH3xkV9iI{B*Nbk|ED1R
zzSXU5Pyzr*)F1$)zXkd4BE0bSv|m?izWO{?KHrB2XBtBOb4zmSuB4fwEdL%+=^_^o
zm1*mwZ6p*AE2*g%2%C|;L#~_|39}(tO!3=bs!&iN6C8-k(o^h%t<TH<746a|ZXvSS
zV$g-u1I(XBaYOsS-jVll&HDZ2b~7UcKHRh-q>Cp}iqsr&)>K&W&)Drm)2)`5CHN}*
z5?Un|mu9vlvggsQz0c+dRftXv(xxCQXN%q}9L$yV{i_&tBzBp{0)z4@mI%f2g-(A`
zs(E>g@e%20N=0GFg@|l2Z9f})W{}MS>HdyDa`*xRji?lbvgF59BW(H#TW13B5<=aD
ztU6BSBty7dHxEN4w%dt#1@U67YrEah?{jju1#ea5Eb=VwUpunKG}8;TyvK4#kMv=1
zqv0@4pY9};4Bg*CD%(!3^^<7wXRR4r+RHPt)-he#UQ8E%9$Xjx<fyd7LoTep5(`v?
zVL!>43NobAM+&scXuVA|mmw2Y2$J26g>;7M?EbAQ30LvsI8;2K7-nPP?Kio^qr8FI
zRy2n?O@W=8new)6v+H{^KgRyE^28zXuh*sX@9k-u)NJ1HNmn61C%i9Lyfh`NjmC+-
zb2}Hx@s4)5NNQncQu6kR+=I`T^_DS}WLpUusk1UDj+p<lC;p)d!W`^X`f3i{`YU+W
zObTJDkz-9Jy*E8Y?5E|9V}8KF1Eg{p9)5pQl+!5|dQnCmBJDg<NEujqkO@NUa*B>w
z^1;-5t;a&#^8RGQ&UAsB{<H<gJga{dhtwaHhC<#|n!37`UR4|Qc5|LAj)J#S+2u;b
zsI=V%drH}d<pX+d^jG7b8pT0^(YSmLkWDD1SfbOUDoC>gGPGog%yafWw7-s}w{{ZE
zG)~h4V(ky?M(<aMmQu`3tl$wVJeS1eI$E!mVK?hmU79PEylW%N{MD9MGwV0+;tkeE
z*QqZ#H73v8mMK5>U-y~}?z-#quu{IWmKl-rcazIC@IS^((YuR7yInDkcFTp%hHUe8
zO3xFs)xnQW*t0=@<Hl-i`!z1-I^Y-AV@d(ldjBR=+&YJ#!^R1z!-fUjs`LS^Rg;cw
z_(06gHJjX49yh<AriIB(=$Wgnp(#U8_L?Vc4v~j493m`^D^2Di!lPfq@DvtMjhLkY
zECvm73+vbz?y9qkS+H4r5vN7iHPv<W;ppqG|BBa1Sz<Xypmu5Gma`D1D?z^bnoCis
z;32%p_%X`RsXbG6HyCdWNi$t}<9qJ+{!1P^8keJl(|6kex4<m}DMhO+`;<aON;uf-
z9M13|bBPyn8LuBF;R65G@`*@^-bRUjOPxm5(Ei)KeeVqD4i&;xZh_J?hqw%Br22O)
z&ajBz@Mwny#PA!^^k!v(Od&Y4(jyMWKP0fs!WMk>Z^@gv*jHkQl&*wVKV7Z5K8bkj
zfUksN(*;l468bxe)$RqJ58~-M=J4abpUVx2+TxPD0xyVhWv)G`i1+~)rY(59gx(3D
zy93+_nWA&Fr1i&WvUinpq)1Ul_j5c(F+Oua{eX(~q0TjiXo<PME+B>?Ao5j#ancqP
z@**A?gAKsu#Q_&8b+QWJEFcbpPfmj#Rw23&<1K>p0Ej3w2$2em=CF%5H;i{^+xT|5
zy%gDPMNHL>Hlb}@Rhr5i11kRjz=Qktxnnl8k+Ct#0tnBi@t-&G3j8tI1}jXiv8zgD
z@7gNM)(}M*cB^t+iYD*;0lac{v(R$Lx{V#<abKtHTPU+V=$mQ^_uZ)t*9`2N>JQ@p
z&CBf9q>OJ7XnRoK)rmHTUBkPP9@`HTC$cP7s11k2uR6zDcUXVR{D|Az(+IO5vTFHQ
zvg@>M`B<~-v}x(GZqdu_AX8Tj9g*9_HjQW91lgW5agYUs{C2=}+VHVx$)5iVt5R*z
z)^FCfU#k)y(Z3ac1H6Ccdfy9VuJOG*x}CYcp2S`(Enh6f-n+<!n14W>>>qU%QI~p)
zlz7hBd5?ELr%VbC*P(olh|kROx5VOqHP)o<Yi^G3J!{8nM$wA#kP)Apb)BM%Vs0`d
zPg#t`v}A^q^N7ZarphJV&#1|K3QetE7g0)XV(#5bkl=e=w&g<ic|!E5=t$suY=2V4
zb1;qkK8WG9!+ud9{+7^n+~2)_m2(U8F*O|7Y1qGAt5`+irg*OYL(EmT+jsMAS?gjc
zr!Cv`=BRu1TYFCprxTxw{L{~mU!0BhoHp{$syxTc%4=_$8)FJ)kf=Ia1*}1ebE`M;
z%oVG9r*nW<(1WhR{==?9WU4Did}<vbGfbT^8rX3L$Obmwd`mQk55y)EhCP#p(-jy0
zx^aeU079Y+Vgi0}^5V16Da(-EP8bmj6s|LqDsad>PP?0Go<2b-HePp%)=&L*mi?dU
z*u5#{kAae2F=M>HQnX}_r)Nm@=A*>mFDk1gE7pU3g_mU}d!U6r-}FB*@-Bp+1|wGN
z=Gc|eS6m=dvNRvpnOKt*^R$-jnePYcjKm+I&$vRtE3F?@kfH^1lmsR1?Nhs=3?lx-
zQFC838Z|F3a@A#QS~QTbwOIUc*i^MH520PO%-0>ObXt7>QB-XOy;c{M@!LJ<{5E2P
zWP&okWtniv(mT(gtgcKi6@+Eq-FgVc`D%zo-d;<k<_tOT6tBjj5{LF2<+4mVuum5k
zhk64>g&2p6)6$=qw1x(#22j(XKYu*&D{xo?|8^u8Bgb$z3FF-OY?8Yns=F4L284#0
zmS#R2ylc0&m*fSX1MKTC^;6b!R=rKB5%lWi=d*3my5d$f(c&m+sE5yVzNCwK9EZ?o
z7>zof0mX%@3wwg+Sy~O2#GPRC2>i*+#ZR`3?w>OxfRe?T+BzCxP!v6ZGuYGwwi9i6
z0B4uZYuVxhJ*5l)2Qog-XC4vGj=o=!l`tw3?)xm6`eb%6^YbOo@*NsX$4Y1f_`p6f
z?oUb0GL6($sTmac<2Vay8ZXA#qri;b*VmQjBpHCo02i--nIhZM56N$Md+c5g?o7cw
zG4}8zPVwyT4ZF-Wv1=L{IhcwalRKCetDyb4(g%3jnvesPm}#XCY)VZ(t^)`9$%4bq
z4HHJfNJ}VMi<qN%+Fl#_>62&a4-JN{uB?bCsZaVk7>SuvI2aYHyt~-mfZf@cd;mU8
zwVpwKQeO$5K=Mf%psj#bG8rWLuqq8&y$Lkgn8VvDyN%Yoc4{@DARG3S0X?L{K44*#
zn6)ba^P9QAe-%<_aoSN4TyH*{)C>z^M=00k#mVv;q|CGAsrX8V-=~$kTD^eLR~v7k
z=qX90Mmr9ge&j%)FM%zCV_ME(S?cKg?ATl|Rcow=*Ea7NOc#-hNa0sI&1m*OdxuUA
z%g>IZ0;u7*vtuNQ8o<D5nDmhLr?C9W61~z_fgm{TtK}y;bq}T5hs%oD5-url$ybi@
zU9PJYrj{KzTd5nO8>#75E%rjcW`HGn4L<W2(7NI|Y@ZeZ$+q7x6Sgs4+tyI=>)d^h
z(b{h3b#-sYYav6fp?G4xvLP*B;@dm!y#ImCduxVqMR2{-&phyo{f@wWuzc%)0~h|g
zJAZ%H|3@mNqxO3y;c7vvI_|N#-uXVa@6cAZX#T`*qQI^2g?DKu!c94Nk{2Gz*$L>T
z#}Sx8D3y4a6i{~*celtmQMM>lH5&s2gShV*x><06u2md}C*bS2r(niMmhLB~$rm1R
zo5M2Qem#8b2-91kOFo)8dLZZcR&a+zlNTG>$LV~zlE<%FSF0I)W&D%UASrr+q&9@F
zisv@;*#wz$$ks`T6?zAuc94$JcLL!U-a*?>l{`>&Xwa6bs%koqc4Zw>Z(O8av2mhK
z+9aiv%nKd0{Y6@bnSWwBz?F@u#_y4-s^c~k)}7uU0mv73m;mcq&rh{D_^Sm^P1-|W
z_E5K)PESuY2YtWTV%bA?@6LVI00bg=*ba1qc%T|GI6as&qaK3c8wd{C9;*31s{Or2
z)aI74BCkB2(4`i>O+9+lUJtCSoF0#KrAg(XSM2b8aK1pFEsSL*km)aQ?jvYk>-6A*
zd~n3@*YC6eDr8_Pq9V_f)~Es-n7APP8<>j5%+!@VrY}RvzlnjI*Trjsz9n0l#`Dwo
zbpwlGUJpCI#Y6Yv@6-ZnWWZ~pBKwu-4qM!MO|h~<drZ@ny!%f-kYV1TFL0z#R*Q>;
zw&||5{p7(Pn}Sxp!@dDEuYeuKPIzgw1iy3Ia=sRsu&d0aV+<|9b-tEe@)Vz!$%PoS
zY)L(pxgd52KgZ+S!7v+iSQ|@6&r)lDkEf{&I~S#h3WLRE6cGWB;nWlXhXq3EGrc`e
z8y)eU7RF0TqYHhMoN#ecFnXz?3325>75&_<9|BZL_E>MB<f9P^l*r9?fY`JtDj#x`
zU7(O%#<jy^pQDDWY*seVciFeZbbupNxt&32N$E<p|A2muCKagxAHZRHo51E~FzT=(
z2){%?(3l6GMF{xg;Xe7oGgA?-s?F+h(ByW)eTMb9*iz{jB>Kht?&1Bbg;)HMe~V9n
z>Ge<disAExi=ZI1gpNtiR)P$ID_NN<hFaWf8tRrQZc~0>dYvU?OumJ*yc&q>-(2Eu
z6a5ILUeZuU@R8iU7T?D*lGi6)f@itH(v-^axXYFa;A7DLk)@S=6M78idndpnYSC$C
z74%5CBWs#{UGh75IsPbm`I)S>W&F4HtB5(M@b|QaMn4P??k*ZynvO)n4);?;mmG7a
zANu5|L?FYNGGl&36sv<-^YnhYUiC*;=~tbR8$0C(zl#9U22@ESvYpm`lg9Nhm<H_%
zTl;CN!zFCebE^RS(U3hPyU@3Rs^}6Q98MS2I))Zud`K7(7wTg03H^>3x9Pa`{3e41
znST7dZ3Bdp)yLQ$m^Gwme$LdmQxhSt-vTqZ@g~qXR7{OXT1s;JW~0V|Cq){H^MEK0
z;aWapTFqB)sHeu9M?z|m<QaSWUQ13Sy-*pwV~ecpI0{_ooeP(amg&GGcZp6s-wID{
zm_m45RT4_BtD(g<m8d-J<3q#)x`gqV?!9*t@?TE~ffB6C23>p%HVFk*OZuohgDxHN
zzyRXl3_@0L>Jn3;T24UAwZiAQ0Pr3kvD#PLnMW6$f+y?SuJmx2C!og@2X?seI>B6#
zFa{yV`*Cj9{$~fG#L<J5$S25Lafp|^yhUqJ{PR)*8=~AtjnT%;3L3I?Mw)VApqG=}
z5&~QsPEM6&R-<)C=wFuL50Jnk(*UO))^hPxBSDO~x!G4{;FCd)B^fJIke%a;QHM7m
zHwP1Pj9oi%vm2$DPCpWmn27`9349y699{m9PT(!$lg_BPN1wf?qB_05CLs6UyE7Hn
zc9y-+?mfGv9ya9r@0W#EANgN?r*YB;$I@r#dyiYpgrD?=zU8(%h5lp0EXizAsS3Bg
zi$pgj?B=@Nv*b`7BGY@OyD@#sz6ai^y0fo*Jy#-&<<-qiL--3eVVv-n4e*><G0!g=
zZ?7#d(p0!zWC=Q)>B2N-dAc*{AA786no2TRMZd0YE|`44QG>MyD+IGr3F9jX<{`X}
zN1RV-a1kR<n2>&^`s}ppb-W8#iQ^ZC*)#2MXXscjv8cSXb^Yi-Ow5<X2Z3-zg{V`Y
zJ{lKQnX0itg?OR_xARc3`&CZ(slU7LMU|qNCdys(XVU98^S$Iv11CCpc!BlulKQL@
zBtA%eCX}WZeD3|tXnm8enD#Oo+Y#Wg)ZrM|Ja&umSL9O57~q*xK<>)V7S7~zY|s~E
zVj9Fb(jb)5Gp(;n>W`g=cSD$3r1i5kj;Ttk8Z}zWnCJP99cTA1kf5+AMC<jUy_aO#
zVoSwJ>62;TgIQXUqzG%#mmrLY_O9$EV^rtHPSJFO0*Q+mA)l5weYjJ0Fcl6E>Iry#
zo{&F48U%iS6=^*q$4NB^CI;R+14OlafHRQZpVR8$k}km~^=qs6ouW3!ErP)<hPDQj
z+4q&}t$nrUHea5{NwdvL;ihH-NRRhdu8?po2Se9o9|V4bJ7{WcS$>)Uo<lN;(B{G<
zc3(sQpVN2u&HDZNVkK6@76WaU`&t|oO#?K87|f&hw&KN0z=R?UQr1|McJsi_Q5wh|
z5{X_-MJ+3fRos5X3CWSvsuE=@Q8|>NY?zghiZk#&t1|xLgpkVJ01@J1P$~Jg@I%7z
zoIA1|l`ohrwv;Z3#d6(gy>d&g&XEEv!ClZHMF;AOaQ5a~DS}4V@@2O>-oz&+IyqlX
zt0~8;ugvPsIm%+cV)Z=jJ`jS)XOF&bknAQBr)e{pG<vm!pS{X2Wz<cPXqY$9PL$7W
zde&p%=G3nIHB`zVR=D&O$5JM3pfDsdMvdD_Qw!`muK-A+bDOIaByI9tT*nf*N^%8a
zDXCV|n)YWQg}&MYYSk@tH%KPcL<@y5Dj}uoYW2!_d3r%mDiiYbW>&amhEw|lFw*b{
zQlmc(euAKq8+f9@b4{wvJEFftSb@1sfElNPFuej{7SX*I5C>|lSDXf@VA-VLQwXEe
zl;KdrBZxE9xb-|7!d2}uRQqQXea!Jj5YezGaaE+1cN24qD`afecgT|o`=xZ}*=q7b
z3>CNhMe#Kb{!ye4UH?g<zn$nOo(>l_h=)M;c){?9$QJMX7Nw%G1En~0RJ1(*(Cc^u
zj=C`U3`Ae|6jAfdYgF<&q=v<7*zzX7q~(?<9*<-MlQ##A=$cvVhMAO#K?uxFm&L)i
zv3fh-rZ2*I&>%-ZgfOwa5|TTC)n*-J4R{k|r2twpE#=3E8X8Rin(%Ulg{Lt|clvg_
z+M2-l2%Ir#m`~b?7tlgj1UAK6yVq(}F24c`XY!a?{T_E|2?3HNV7V0E!-m~e6O)C}
zZ2l<jLIbs66?d(!w9^^|op15%9`x)bT3;;_K@d-b%Fjf&1y9X9C&G85r#uioj6~g-
z-PduttlTXXyLDO_#7lXhixVO#T#a=hOqn$)vjS4%M~FMbU=wk0r<5&;(JH6Qv}hOM
z*yP~2Y+@J*zHF_fRrv%)Zjp9#USz=up8}=QU*v;1)Th0@GZ}GBnAc)R08`YaWC=Fd
zVk>c*;3x2?3mr=?zSd-+7)l^+PYmk$YXiU`2(XQyID@c*ho$!iF!+&-FNxV=9<SUK
z1wJPUf<`-DYsCH@hr`uVE>;@-9O-#(vfeAV+2xC3fzmncT7$AA{^)%bspzknz(^>V
zL|PlB+;R^O1~?z-=|Y$h1G#W8AgVyDhbF3ExQ85}#?A36qnf7H;oyUUsD}=9fHFbT
zo2V43yMj@I77jRGj564~r1YWRJm)u=$3Tvg3OgmoJDVW-d7G${B@8_qh636Mx*A}7
zhAJ_)b0q1|L#XOtBzP%_cCH8{QI(FpTjf5}Ocw!Nbyf~y2{n6+5RpH{nMw%u0cB)y
zoZ76pvb92qK6bXQZLz8wzj)xy+)#p94>#%`hs)vL`eW2zER`m9GJqx42`}jQ3Cd6o
z3`(2mJmn-#8O1h2aC6RySMpI^UHZ9-X?lpC56J7w0}8akky6UAbr}8nIZlwj?iz`z
zA<L9rq|~?eJ!qugi6$)=CD?lS)$uU~Rsx0rX_`b^@&T!({WiG`c0?qY1FwFCaB$k6
z=Ahb&+vZ9R!xgz%k%SS8_HI+zNobfgASE>wtcoihBv<|bDlLGAFqinD09-WQ2C1B}
zElFg%0l{;bm>kL}73Z0a`fL7hFtgV;hpR*>Y@kcd>1&@{bqe|*n>C*Tj`*x71`E)+
z)}nyb>S0|QgpF50QAVqo>CLRbbRPNU#1Ja8u>dP0PldEvi>e0H%Z+s=Yc>=Jg;jQd
z&qAo%=D+IRg&tv9de7!`^1@q4jbHHu8*sS|fI&*(AxF_;Fzg9Rr_|v^5jy}G2A@P9
zX5bebxA1{iw4pLd?e+7Re${NJ#Z0v^=*_GF+s~i@I3*EUTgjTLDwG0%3rR$^Bpxzf
zF?7LikX$!9)LMQQbaqZRqmz~rmMj|#gT6I{W3w68owZ0NzKCg5B;qt&G7%J>ippXH
zah$WBkEk}6_TGe&MUFUj^TPzE5$mA~eX$KRr`%H!ph}X#wmhyyUjD5!4Od`4VG(Pl
z-QX~Ekw$W8b^3fVx8@oab+}(s53$wX7zcgN)JY-KkG)biKvfA1{$%sKc~&2LyyH8&
zbiC2T(m$vSZaOa<105=ax|Wi8e1f-%7|hC0cNolyfYL5d2WbT>R3KWR$(f=xQ`M-p
zViL9&f&b6#>sGB`9IeR;CA4qsljVjdl-&*Z`BE!j^x7-SP^<lPxjKfyCj&<70Uqtf
zFPq9+FHg@mb5d3Uy?u3eu%u9yT5#oPV(^Vx`@#3kviZl$v;Lw@Az>$&_mFCN7MMZz
z6o~4jy&mv{j|)5JOtyqf1e&Dv-803A>x9{YJo2d3c3H?3QwjNbO4A+~&ZtTtPQnGc
zf`i2`P}+<a)dw0MHplxH7gfBc_gCIZZ?GvMBt<=Z&7)0oV$~HU#aW3O{kNcZcDP8Q
z5(cVVGJ?NlGzzj!+zq3#ZF<7{7RTG}fSVa>!kbz|7wp#H%XA-5uAD4Fv-8#p8)B2L
z9D$h0h0<wHq2`3iy@i3#xW`mw^kJ7=g;B9lRhN1?qLy0Au%Kso=7(5K_d<V*P-jbB
z4XZA90yiA-NC-}P>JU{5KmqW`W*|3(Q}UiofrBHinH0(@o!nJqaLt--*}oRA(>ph(
z-NpweJG15<j^wbA6{piBTwg8#;X(BfS|Tzqlu@X42gBbVarfC~%g*J=<~9Aem0HVG
z*wY&nT1s6c$(aD(8y#?tvPb>8SP&${BN|0zPI(v~En|yxE#wl%5=D>=LfsDr`p(dE
zZjB(Iep!_=)zwhglj?+0?PdgKRjb9>fg0|O2C;$GQV2KC1WrMsL<yo_ce%q&RPmi<
z?&$6wTtSL65u&8MesH>P6?klk8gNi{)koO@pUnDwq$`TA5JSGUS0FXpZgKYuqM`~S
zpsx2MF+;0%#8UnEwxZ_Rs65Cfh((!!ie}3&SF1|os6B-J`+!|BF&nI1gN<5ib3+#J
zHrqoWes7L*|9H}%S8bH+s^wG;x><(u6GgD8v&Ii}HZiwUon^5aUl{CBn^43gXa>90
zQhEE#PeUYCQd~cbYmzM@UWAM>6nec#DbqUIohyB6C~*|^U|d`*1+5y>B-zSH#{LyB
zE{A6g)o4Yy=@4g&{Hwa|?b3UW<pe6qwe*%l1uZ|UDX|^iB5q`)!Dr{owk7HEb10am
z95961Y=|`bbNFeYgdrj!*}=!Y3rHeTLLvVMTkuODR6sV-iNKTbkc=w=tkSaPl0qhd
z`X>+$CxcR{oW0l@{qi)I?+#<IlLjba0DH{X@!Avcn$0*0XJI_$N^}`m?k%{Mv;5AZ
z_n3?oe>*REJ6FG0LcpU)<J}t`r!E5m6C9dij3Xt5QS;9yj1Po6M419Q$i9W@v;sUd
zilhSd7D+_?j+s&;DaRaBA#<^1Z3}Ii7I>`L24sk&(~zbldZTsGxM8Y>HJsr`{9*~n
zta2V%y*=q!1w{;XsSFQ9cZfcX|09J6Hez-X?-<ioA2OLc)W%pM3^ohmFq5<!n(8m1
zhzxa8^ar;nR2;%?Hf^O34d&-!0E6)Hj2F|W9wFm5Eh`&4`y)gJW0_J#Ep39r5Lx%O
zO!-LI0#R@aBu3h1XtLbFw3+-yW%%6SVrtZiZAFGCaFQWl5;1D1g)p5!`(){7ArFDb
zyAg{nf7-s0c+-OsU{>Pdp_iqS^|RnTLSF#ul<yhoKu~^cpCZCl+nMIJfi|ge|Gm%>
zs~(XN1X4RxkXnKQal8O^1?f=gZ_mvJ&IQ|~Ir+`lPux=rpj-tO)Brp0xD1%GUvU}S
z$&#VU*%_Wx0&dd(bIa4l>atDG@q5)}Dfz4l-kfu2@v+aWo%f>Md*Ja#m!ak_S7qkf
zfDt0*ll3YHJI5H|J}qBDE1tx5*jl$UdM{Y84=k-R>$M}*X3JK1%eKTr5(R>o7aLuw
zwv!xv9zl3C12m&&STKAomXpYcQ|eHUMOn9456!DH=mT>RIUsuOz<@1nn$3cMi{dA8
zcuLB3uHVE6K^OVO5y3d7&6b6sO!t1lrlH=wVS-Qv@p$Jd9$F5-b8^$+Y&`Vq4@|zD
zo%qDs>qR&UpiUGNnCkeuN7}R5p7Px6X$0;mS0K|iJ|RXMDn6@r?`4}WSem4CNnAEV
z;Zbb5*t;=ChMSU<A|&vdmenuV*hzmp5(Dle!_z{fqgQwr@_<yX!N?daTZ#2b7|Ct2
z&rUOEK-uFQcTR&qy&Hdnt=4He%o7(1#jZ+<jxv=ms~$4ofW#y7o)k9Aqc*#5ABTH{
z4z^>K$O1p)`e)|w=y*Wv043Jj&DCV3eRDbDieiAM=UY2hoVMCfPS#0<g3`=aEa*DE
z&hpod$LZm5nh;G^C-_7Af#eX2`aKf2yxHmm*!2p<51;D9&hiiMs=`}HU`Hk+3m-4U
z{V%+RVT>DR39wocP#huM9bj3^YHvCPiX`iOVG9KOItw$qm@Bh@Zs%-h$IbcBEVtn-
z33Nn2(t$}I4EHv10J;{$$BUOX#BF6{>c!IZ#Zt?~QlS6+!V1k^7O4J>$)UI&e8al2
z0xaK;1v}o5%YfM;q*tiK_t62~Lc`)hUup%C{KmZ)B>UYl)s&Q+o(N<VM)dgfgCT8&
z?>-@6L1nwQECyGU=Xh|uA_1KRwclNA+y-Tzw`NtC1l*--Jy6+V`;KOf*<({~)e)PL
zD#^cgxpKC8ao8Vev_dOOzZK`5`rB?G(BO+GE95LM@x6R+UcNtx9fqNRatn*eVMgi@
zzK1tPyWS?g-F|9TfOniY8`!Y)5!@}L)SZG_jxIUVxm1Y^Aw3uw>3akzLM|Oi0F<40
zj)-pOh;n}e@KVo+Lf>Y|i-HB9pv3W-`I~}ic;?XP{IZdxPzbM`Bn<^sq#)tDvgfw>
zg#$mUnw<)(5~jAIqB?w;%wosaDqELW!KjWondc#o7dQYE&aBz$6rL!V`?c*#MSQnQ
z2bmHE774?gOe-@$YPAxjDjIGez}Lo-_;cCUu_lATF_9LAf~Vv+Uqr#jJ~6`QNA&jD
z=60*4;G;cod_M7<&0b30=_T~x)t~LN8~4n2BnW}_0Jp5K{<%$(&!FX_EV)b~VDKD*
z31>epO#ek#fK-?A)w+sJ7i5Y(?a_m<rx+Qp$`4*;)^z>^+IF`d4<cV~gYKeW8E#BL
zZ4AyeYUS_MOUukaAAFD=GXR@z8FmJTla0@?7?I)Pv=U-h@~<)446|}QJ4IUC0dLLH
z+jut`QBJPVDB%Xun*OPA_J)kqzEZ?chB{c}MO(;Mk#Z-^IMec&V0?L|W3@TK{1)xK
zcCYEYY-u&nUL6Z%CT(q(A~E8!y;V3s_93<r{ylZL<dyDT$DZ69+Nvw|px|p8GG%W?
zK}%`4wv?beCGW0FGRH{TbTdxq<k@Ei9XP>Yi0_6H(CWgm<1RO=zh0HodiNG^jLItv
zl~14q@hW|(#h(vr$=$sf+=-6_m5<QjF4`)?3GhIT8_&7$>geoj5+~YvEqq#aO8nqa
zu7iE#wJG;(k#w$+%Jsko6{CPzMwn?X-o+MUoQ$gjuT~S`u(!84V88QRmckiV9cFY+
z7`Ulc3T01~GG*2T;DZj6?!EvW@QFrTEiNkgRN&Gupfjbz?!ey~B{;kTzBV-0D9PIr
z{qvBJri%7XUoN+X*V_+G^$h?^ibkCcy43XaCg_1bf`mjIm`DQdpk^(PhTYD>S0sjU
z#yV_U^zsw(EX8<1mH+IwxXcCwqSl#z`O3FH;`q2;n|v?E>Y+dI`QJh?0zE<|sGFf~
zg(7?qCe$-Q5!S1&2Bz?R`W|fvNVs>`-OdypERdijqs)oIRSby2QDUV?)KVo%<aPCo
zv>qngkr@Rd`@5cRzR|_QBG!*82)TtqO#+@NvWmMz%zU7pK%WjeGYHJ5ur?$RVG<WQ
zp4ON#Pks{k>-&8j;7^D28!PBVKxJSbITr<=9NEgG+lU~}HXQ4%3&71D1V?HAv-xJ8
z35E-|kI-FE7NYctEIl|M*4~gboec&iB#0G!P=2sR?68PQS-^aZ*2Q>6f&c6=_7t=T
z=6O)fk*^<K6vVlp&tl^W#!2HC>y5&a{V`^%12u65k`nmC3*+bE(jSmWy>QjkthPZm
ziI2toD)!z#Nj-?5!VdXQ<{hxT_uT_?H?Wui29R=W>Y}!g4;k*FH3OOL1QFPQebnb(
zTQgRd5M~fQlhHj_o3<Y|lDAZ`AYE21J_2@VvIDVq<)!ER`(g%S3`qW1uW5AxS9AMV
z+Sg05JZ&EjA~8TPNHhbZ201tNh|C5=0l^$#I<8&lM*fQyWNO3sI*%_2d3cMY8RaP5
z=v^)Li#5kuUwb=2tRolwb8B3CO}C=d*c!nGXuu%w;P7Zpshd7If+LzoR~y?e3BF#t
zJ~r>5&BQAaiBl^3rr*pAHL+|4sdmcmn$SGkWE7R-*Ms$4ZTZ}H!9a7`O=aRt7w{xn
z!MG9AtPO_irzL9LV3)R%>ZaT^J?{{)K&|beTTi{JI;~Zh!=YWZ52_PeO^;{j><NP&
z>O`yffB3^05uu!koGlxUu$`vJ#8q;o4I=yR_5g#1;H=zSJS1HV0Alg$-(cJh>Tn$G
zd9ke!>iLPdriA_kF;->Q#_b&P>#kBYUv|5|1h+}k-tXr;;mp=P5w#e9eD6tQ6p{`8
zFyA?+2097ZdS`Kgb$VbS6_z-VewJZNeh@X~k{}KJ+E?+%HS2N50Expcu)V$_*3_U~
zi<HBp#pW;y%Y?*ru1AN%bjn}V(iulZrw>D)j7i7l6-;Zx^>LH>lQaGK@Tv6)k(Z54
zS!MqO{Rby|<B6S)r~A)0y6>ZUKaP-iQ3hW?X+J}%+(S$IU3+CS=%3aO`FTY2oi}(2
zS6l4eqn<qFt9Xyl>^~;_kB0~0UCbc8`^)aXA4R$>+KBxXa4=@$`x{4vvse=?&IWQc
z;0)gJP0tDc$Hpk+!``c1IN<d6g$;2ySBTT5QWkk^UD%WXP~*qgy7mJs8tl7Yx%8%D
zT}X$ro;$_IuRFq8&H<L`4R$=7XIB1af^!b;rvP{$BZQlqZuBi)is^3{9ZP|@w^>Lv
zeT?*>p=@kyq5;#(x>Z|o6!>63Pj?Igm-xuyd;L+J&ueENe$UWHXQfAH!O+06anGkt
zPgof>2(hitT1FugAzI2=)k}ogPr40|f~LtDxfdZ8?d%&tJhxI<bs1t7_?ibXCf^UM
ziW4P$DnmY=2T``G-!zk?53aUht#}ISkF`p;oLrqTAQ^)Y`w-lFY&&v1E;32X`do=~
zYdC4QoF)jEv;)6ip6aV#vu6^0Jkz$w=KRp-hlb6_YtyW&75Xu5o)Qi1SUw<SfR>KU
zXBZ4jgIcT5BLoBkMw5gOGq|l4^~N9@OL=)mAt;<3X7$~#BCrHw5zz_-iNGoUiw=ZX
zQaI%)E=Ffi4|Cc$E|KGTEUh!+_x)JHIdfn)aRD}XcifGE8!Zvhy~m1D^6tIBq~{wv
zLRK4XIsTPxjk1c=UlEicpcej*1^(Y7l1_eS8O(u|ndLOo8fOvGejxo0eQ*xkz-h^w
z)4Jh5<zS#RP3OK#cCgbVk7-KbeJvDKa%~Y(Cm=1u-EUNKBun2;i4#3@=ZMyTUM05H
z9i|VG>g%JYUieln;9<}_9ygz|j#&r!D#51zc4sRpPsCe{{|#pTwzY%rj@%AfG`c-Q
z<f<0=&i+a8pP`UJYzy5!AOHX&Iskz1cPON@i-(P=^Ix%~hZ^fnYYd3*XtCQk#>S9_
zLQF~Hq|E7TiR1U0BTN=X($%D)Gc`3P<KSLYpqVVPIRfpsH`%=H7K6`+(wLajVU9AR
zJ&p2O#r}|m_~du~&rb~RbIfpbfExh;DD1Rv1C!&CccwX!s(!6VeII`^6~N8>@9Aup
zHP-Ov<4S`bg*g~7ghaf@_Z<9%dgLvF!dzaGf(1s%ergjL$UEMH%`pH2PdoSxjyY*n
z*@5>z2<COg5@9Q2hK7Y5ctqw25G(azaO`2NIT19Hnd%Ktie!$<1PMl90LwUGuG0bq
zV=|4E(aVuy;VA->5wvKJwlHXbxadLbm;k8X@q>*I0yz+MXjf_Hr8}kgijyGiXkPt<
zqacoW<Cq2$dg~FMmym(zLcTc4zLCV!ka9{u5KFGiOST|kh3RQ#le{v-XR;<9MMKn6
z^W<E-=Ae6lSMnuwoV1jzP`E0oHBG|J1XdV2_0WLp1>6J#4RDf4h##Wqm28zH2XjyS
zW0py_z0^6OE7Co2`8j`yq^JfCmG1nx%Q%bjI`|}Mhc}fDQSRClUEK_Pzg@y;O2X8^
z!uuK(^=h?3y{^hFVuT#6=y4Vm7aBE}Z0SmM(TWdq#KQt!&Vv-{CUk@y2<oO*gk2)4
zV|@>a-UpDD^$*fyIojkYZq$SL-tX3dS6#woc-U{921ygrdDhN>j?+q&g+B0b58%#@
zKRqKDztz;aT}61~8GB<o5bfp)*l0WaHU=V>Hyw1AUtIrVm=Rif6yRbUWsg-XBd4>v
z)xdl~1aqE803(7=re;gj$hG+(K_KI3roycJ<ML9@opqk7C_usgGA=d^S7Kbez2-Y6
z&GsCsYmAUrFfT$pUV5B!117k8c6ao!5;#0o%m8C&e4THcs-C<K<yEzjGHd)QyP-JK
z#qcrFoU$Tl?n`HUVljPkl4{<`h&qYu!a-!a#wL;gO{rXrPX~)OXH^7A#i9uB;X<V5
zBpaqxXx-G*)}c!3$xy@v?L0HlR!%!y5e}bmgXLv%XUeRPYlW@fhk2HoiJnZ^1N)r$
z&pGc(w$}>wjyA2jc1iz<IC(53bxzs1AtAKMN$0qn(`3^-@J|XGc=JId=Q(`0;+!Kn
zt#1(xWD!<-*ox<|gkvjNQRC_pzJJhlDWvSBQQFlj!tan+IN$6!sw5wx_20n%9;*Lq
zC^^*EQ2h#6R&cU+`0prjBnXO9uCFNZK;W+c=YOweZ0}_H*O*RD(6tX>Kn?kX{>_^d
zz?=vmU*B7lNs(zV{VwadV0KSpOYw50u7JdB@3z_{pVM(Hz{x*Cz82^zQ~@1NDvCr|
z{7YdSx=KCr2}yKHdC;-|P2~g<e)23KJ-`yiUJa*K&o44S{8vCa^hF0d*ig)cfKn=`
zP_RQmRUsel1v<`H*Kc%(lwxI?US&Dn!7K*0K%}hXOgO1K<DYsxSRvq~lvfOrkX`lo
zl`h{2?nlFt`Uv1QMCo@02r@Bynr#!=u4ZR)cBT){Jo&9?R9dLOOX)bjrGcF`RV`Di
zsQ}a8oYZR@2xsYs$l8Nmm+XhV{&Z~(?mX#IlKo-Ck#d|TZ?~gJshug-PD=26H*V#B
zN;6sP$Q3^NB^V<nsyUufa|C_eCY)ha>S?kM+pL&M49QLQ?svyi{lE3`&nR-u?-=mO
zUqar82LO=&Hl22^wnnB-mUiZUNqSLzJZ7E&^*wn6dAw|&#hWr?0umFBdXJ3J%kO7v
zl`22+x@0l;1I<b=`i-R*l-pV%!puxDEcCbKuE%UouXE2$9ldSmBQz09YEf8)X$X9z
zaGA2S;hp}?i$P5HHG_B~a<K5;2nz>qS4}bEW99tHpTmh3Oc+YLR1NTC(4EL|9ZbSR
z*EDT2is@fFgP4h;h^Hok980`%C}d5U>3YPRNex)CkTgY7hC3wPFUeHth*h>R#~P>_
z%MNeqEa^yw(adEWY;NU8amGy(4KR_}hQcLDiVgKEAGIOzj3dcmY={_eL>S0bP_`jl
zLU@Va@tjB!Gf@o7qZE+LI3@BhSk~<dXfs&ptm5|aWKlD1V*$a(S-sJHMR?7A!!z>B
zf|*t6{83?VByt0SEJ(u&E7O|HyJ-=&duns%VUge^K%k+3)whm*k35rX+u0Kj3zRyW
zTAkS(zb4Lu&8pvbBOUrXQGjKqYZaVelsod{tu32`wIwi}^>Ea>O{rgluHS!fF^ljj
zgVG!Cs8jNohjbBx8Gf&b(#Qp79(Q0>W%0hSnOFt5bX6KG`LFegszXK53TuypIZ|cG
zC)b1so^h<>+Iuu`Ok$A~3wDq^M*kjOo}PReFVr(NX?C36qGDHw*ZQd)(@W_flMjY|
zS1$P&+I1I<W3FL{mmnT1{4rm4porOQK)?h2O1~{<N|0*^^Y9*{lWD+X*UNW4GFfM$
zT}$9{a1`{;rpo<wjrC+(?zVvTQ1Eqm3HfU1uCv(JiE$G<ZF=p?mrBr`Q@L%pe*GEZ
z^?o^)nh@kP;Zw1VvhZsD5-!EA_{q&3=aPVL)0f}Tw4JAhUQZ&_aEQ0g27MW<>b#rK
zpEW&Sk@Gi*|5u{|{srRyx4aXAf`pW#ukWV5!2VZ)%0KJ;GtDGnS_*^_F;t&yqLX`F
zO{CD0Ug<=(kWWCu?0euO0l9LOy**`w+Uc5bZ2H^LIJ+=<h44z5`re-M0)wrV0;;VO
z-vf4b%h@PiWlAK)8dh02)02*hZMbMZVZs$sBpzkjT%k<emP>eG{Eq|thC2FUip-`9
z+FN<>Os2&l%bn<-0c&06Z9!RWQ`$cy+z2Sg(kXn%s8=22^LJ_Mgda8SsL`)GCeNLI
zc2f)r7-b>_vN6gsE!&~XA8MDj@hmxUk`DzPE$blQ1?7{lSDQHDGq?RJj^D(P+|5D5
zW7d~)zhzAQFUmCc{jJ6JOAMZ0C(hUTkCdK&%EZFd(8Sb<@n2auvQdKYz(la3H9~FQ
zBj}<S`zP$+4ekMOYJ~8?3PwC#G`b%hj~^XAcXoUmENG4nnKYQ_`q`2Jfy3V1ItIE&
zI9Cgyd5o&8BuK7^o;B(sf8v>ct5A;<*32>eVbt!C`jGX%F}c86|Ag4N1^HXC|Flm3
zKfN@ww|Du!q?gPWrzvt@utNKCXo!Ey^53Lbl(HW4#Xm#z$;Noog8JtmH5^a!leH{W
z*B}^Th%Sd9Rzqq%7eh}_9we8hkR+O*cZb~0C)ht%m>tPuDX_W4Y0!3UPJe7>pBaR>
zORV^ue!M&{V713KNFs~mxlB~ZfQ1Vvv(IUP0w_@pp;c3nNq_BeZ;ECDcda7@3HF&n
zzGW{<iwijgB{OZ1nTVHe;EJEI&@nJaa??>Q94TG`1>gQ5<O-q7CD%qqc{f|>RP^6f
zek9b4BjG9{v_>`wxUH^@lYKKI{}uy}%AEBB6nR(-#GINqAG&~cl12>d;7Jov9J~Dx
z$vyCO%7Dz!s115W#3v1FmP7f2bM}mt11XY{HbbKIUXwB}u6E3^qKb1}$KD;MryD%%
z-8-})<&BUU`&aD4gq}2dKYQJ+Fdu^KCWRS_prRTf7i-|5hq(ucA2B5>pDOH?*$AQ+
z&_36QJrEPniyq!$Hx?_fnkw7^Lu!qHm}>L_Ls^aFCzRbn%)zoG!iB^;m*qx`igd{U
z4iWNJ%D!kkz}RC*v1KtsvKY5us}#wEUN@eNp6LSJaLxYLQYM$AL(t7a6D!)!04d-B
zW4r3cLnGF=(^aTHe6A3`1q6aFFek`!f#1czL=wCUBNr#Mmk#zRrW%>(g@ePMF>P~$
z)<LAR$|J>Lle518(~}o)Dx0hI5G&UOzs~OU5<J#}{n4BFv2SEnMt3YwCbPatw}*oK
zjV5*_%UhdH?`7)rb?Eu=xaQ8d%PqSstgh2}K5|`?Kq8D*+`N(IwvqLdv*Ew8z%q0X
zMUlU-$MEmi`?s-inDXVk{=y!ZV-9A1GfrLtb!hWiq^_#e3m}HESKBLY*uPo6o6R8M
zoLX_sg1)@7*No?GlHVA|(erU^JU1UZ4b>78p@1PbU?eJB7MrgIAAJAX&Pki%*Q~!Q
zTHHcJp`l@v6e6&aLX6=E)-1A#U5JPi>0KsLgjE?cOT=DjIu$JxSnUQUI$B6=|4p~0
zlA<Aoa~UVwgQ`8^@QY~$cuIAxp|u**o!RPa^zaDFhE=>vm>~T~Flkk;-8IrnW{2&F
zPy%g99}8A|anT|n7^=+0*b~hp9Xh^J7Iprf>59Fem+LA%4NLV!El4&AzjdJdDmA}s
zQ~|Tw0MSdBGb!%Ty%dL;QR<U&)vC3f*E?g+iM-#MPBUk$mfd2^5M6;c20gFz3_@L-
z>=r#79ZCaDZGPTsZt2lY;ZMU9)`EEI(nRP~%acrLPxZ5)5v8?ut{mFxD`RuNN0tsN
za%hy(BMYZUw<P@m;jm6qmDfWy_1BXl1SQPcElkJ;TRo%bW$js$;6U|=UK#&g5MxHP
zT8v=add0K5?TN1FA;osU)fRSarzw0W+rl^I>fJ7ZA}R&l)h^y$IJnH8mRa3$f*$FU
zFy{iBHh+q3EUnuNb&1J)PSCu0)n4vzwr+m3ps3`Ir5ISi@a>6+9+CNEvEf_F5#D{?
zYrJ`V?8*(@ZsWl%cvUFe!LU5T{ilUhR!~Q&{vz5OU-&`!vR&H3_I55`Ihgt?9uB6?
zx_>!=CbeF>bq3gXwEkrxwYGi*SjitNOZ#iWk3bP3+-o$7aVL%MuS#x4J-V4@)~?LY
zPfyt*(ATF&qs@c(LM$gshpJ69Fb>d7j+1R~e;nFd>Fn7mLd_BCc9bxd4`W7r(TL0O
zG_bzi?_icOYDnUSh`E=2jDQauAKpePoIv8zQvRW2ueeeeuP)kKkCU%jE~+crYA>eo
z;c@&?n`q>ia<kr_G)M7~eur*q{T>v*cNaCZfFc$#&Aiw3Jn+b2=xCQh)@dJ|Fc55X
zbzi|r1J7Iv+~_U4fkp6%Y`~&dc%6~9_vsSiYKv;2l4U9tyxtb61NV~hQ$|Dq<<GL)
zFb;N9$m0o*(F|X10H((;TkYtu<RLhAXL^wLq!iCc6}q8>e#eHVu=@=m9$SN+Eizvn
z&|n~RnscH!&b(ppE39Z@G@k>1RSc*L(>d?o;t|4Y<Zv&<9(Gx-bJY#nkZ;~UR!A}z
zMWQ9(E$#NjI4YpBN3jsZS*{vDh(9ts-62>iz=GL;n9*Y_j%;_toYwm`*l>3$$=wK9
zgXZDE&#7M%)8+|k74`4Cg??iEr+PI!)`=6o)OzrB{-xf3O}C6)on7p0|59((lpu5;
z17h%QMbhfJ3I=hJCHvl*s4KyoC60xe5IOC`gBU2n&h1VAuytFiegs?r2RYnueE{(2
z?T*FV^@)8b3~B-_oSC5z95=L!t#d3c19_;DiJwS+rc!A#GIs&H(vi%sDBMm-k0Hq$
z4z5V`<_dACQx#QnQ+3I6gx%rZ<%*Vjdd}MSjzshB(DSl~|3c{ATh?0QSJ(d{Z-05$
zfBgkR2Z#Ucf6%#YIZXUqBaY)mejCDQBOD7e5sI{@CrJQ=!}HVbe$zWTdYzDweCc6b
zc@d!TfJaV?FJyg}7|c9yiY6%c5T53>x|1Rzt{ws%Okl}rijz&ME{h1WT53p1Ose;Q
zX}-_`w=?+LsybleRAt#5OdH%~2mS_PtY)S1S$`q^{cZoac6hJs^1uE-X836Q)YlNJ
z`l6=)nqvKBqV2vI%s&>W7QgbG_Zbj}zw(?rU6p{eL}fYs1U3*iy#f0i(ZLsWHWYdG
z-H8LTZ8YW3ypG}BW<Ga0=__aJ->VYTIOuy(NwtefkwS#2CaVd3efVj7;s+qiq>6jO
z0?vaeT7J5wssnlj5cvcHZoWLJ;3=YwT-kBc%mjVY0X&b2;i#Z;DfKAWMNhIzve|Ej
z2Kd%%{QgbGc_w*cxlehJZ?O1kJ1j7*F5Knmp4R8Uzf-`Tsgq$nO-k4yU{MRd7)sb6
zY3cV-aY088_g3e<7`n$E99kJnVkHVxE5N?B<uOZ_bbtQ~z*3%U)=IP(-9$b0&(2mI
zd?=QC%$t+ToRx}CMp}`%OQE<7{;|Xp`;lO{+1!MGw`xYgp4qZKNF^We`DOhP50}Nc
zOk(P6_Qd+YDg6V6M&0Oe+4J~-jkVbgxK%d)1uVxyxdKu%O>a5HyS;2oPG;eF|5ko5
zMhYi-57I`0sUZ5Qfko(a!x@A<)-a?+PdVYWTS;eKg7;KKh2_oMJ&`7g-^^r&+~t=;
zZ!xTSIiXbs^9Zxj6CuOxN+;$L+qdG2c}1~H;KiE}WX|g3lT8hTnq%qz*6^P8;{}E3
zlO<4Dl|$=k^1mk7|4w1?zTo?BMEgH5@Q<0u|3RwFFBKEr`-RekFO>c(o!r#U<X<42
zOj@_wC;Va(gL{0vF>Q-u5%ZDwYMm7sn-IB}!=*ihJr740-@PEGnpKn7*gF|Z((OOe
z87u8Stdbl1JfMmRSDDDd8E~a6-N2sjUJY*&H*A)4>0~SkA#-v&|F@C3+cMEOkqA|?
zJ>LTz7>dR*2~4*cV|Ap5(=O0vD%d4K){npX`wtKZm}TZD#Q-t%sqU@f3=1yL2B)KV
zD)&w_Iyo#-845J3-%V4aGq!%wuA#SIV@ZNV6LB5b>Y^!ct8-<)A$T@PH*QuVc5)86
z*>5bo-$=9jxKN+NJe0FC8JyRtQqC$H=flT<iWWe7dnIQ#E{j|7=(t7f5o>+a6RNe<
zt)tZk6XSY@W`}>qaV$4TOedvE5Uq6WspO}2VgUxl@7vlIi=AN87%&IW78A`lQMQ%U
z-MfRAUhN@90Cp`kWXFc3eWgjx?I|@~_D?=yYhc7S&G{2PpW<h>b5(*)({jScf;4kQ
zsL-o|v^R2XB>9pH$jJPCAUxFhlY7N-T$#9HEc~>K^~U+DoIjy2@+t<|uao@$=^6eV
zm;dz~hJV-k|7lpt=8TQyeZkWC>v0kPf~B*mi_6#Q)IV%zQPSpL>?Bx!aK@XsT_du$
zRNLj-(jTK0(G~~;vQ>0q#Q@Uv@%7l4{Z@W~coy{p!5ePz!ojn3-zAeyXhrTyjk+n#
zVg-(`^vjiPettE5O>~r8(0v6}n+oFg&R_UrrAqn2j3j^|r|WzpTdA%Co(MI~FzX6Y
zhuK9-+RVf1>Lvv*2f^FOj*;)&1epqq$S2$e2=Nz=54@8e!KjtGwQ(oREE_^AE<U0z
z2$KBP1@kE@&x=uxXlKAFo&%=_7BW;6!mp-Nki#c(RF85lYKCi6x4Ez$4w%{(sn%X8
zV!du5_XN86Eg^VX3SOKw5RX9MTIf5B4~I!6lLv3cM)3r*@kov%97BE9ldIui(UI3B
zQg`{SgNK%Wku8Aj%;n6`9j@+~dyhu(4P1@=c0T_m``ht~!Mh2brk&qlT8eyZC%`;7
zP9n79rKhV;+6|6-5_p~d7wm{XL#+IrQiq1)PSLjV-e?cy!4ir!*^+qEA{GtujoO*#
zlQi7q=K(+E(Nm+EAyW$V;ga;_${81eQKa~!ZbcL6?{1I&3wDtqoCAYj`W%A%H+>r0
z+y3?3kALcOVbf`k5EFby+mS;(K|`5+zL6s9;@*87CfBcz<qbA>wq0L|Ofoe~E<pU0
zbUQJ1xu<_#k|mOaHRYRTpjWa#Sh5BZ-6N~J?`!8qK(MO>@&cu#wL>vF4=-OMnwpTp
z@AE#8IkhJhkfY3M!p1a3|8pq^n<Q24mfQbl-<`JJF7#0}T=CURXVV|Pq_&>L>KCMc
z=A<y!AI#+u<qVj2P^o|A*PK-?v8y<Cc{kOp^0>9c^M7Q_Of}JsCG*wRCTGr8b6r3A
zfZ!67ssA4NNjA>;pZP67;dA8$Z~p8jKiKnkSXUcsGdJ(}<KyvUd6naulQ|FNxLh+f
zt2qXmT1Z63KA--#MttwCGx7F^T4rwgclyNUL+yFi^=oCMZvD`kvZW{c_Vt;oes=39
z9G@>%DB`<g{nejG-d@sj-nDPj`utlOK5M_q+QqMuIiUG*{on3qlm4U|o>%*$EPK)4
zr}M+#nTnks@*jS9uq4av=`&&e2@E2F0ZCRL5AYxRmgQjc=EAv$!rR>nPThN$@?=S?
zB;z!TLus0>FGR~WpHOVmk?t;uTj6_t?$H&>d-vxm?~y3I(0nZaa$I}z+Nqz#W=#FR
zZQa}4#D4}$G#mQs8Hp}xkYj$Ji<;!Oao1#lNz)lPlZez#%PR><Ov(XII?TOzH}8;w
z2>XX{#XH(tXZB{Uxu@e@w%?Y4X{!t0@`n$uY?oiVE&EaLT~UEg|DV?6d=D>pbDU+<
z3+;VTX>%mk2SnGe`SJbw@ub6&t}nNiyzo7m-}a+!``#(0zdUr7+D=`db??ZHX-*Te
zcs!z#vTJ^=F>L60DcQ%AVlnr2t5Zhm7WUUUdpd1SbNx#=q|J40Yv?oCtCBtawQ($u
zCw0t|x_xOuY}`dXi4y4m$wgNTckE`r^lM_|UR%e6P=`ZSdcns1T~%5~!z1gNzq7A-
z&iGR}K%;nWdX8M1$H|q+Uu~94$gWmD_rGf<!;-~$X_`L`!Y`(#PU&5erZub5b*=k5
zw!1wJQ|G()JIcjOx^yivO6GN9+5DLD4Zn?UTwoI4O-i!|DM$w*(ixdV7;qnM1Z<-L
z0jQ?|0^sIafH!U}ps{2K0BKPMVx$%xLL+266aDZbs7@f|gQ^kv93OBHp__z$`VYdS
zslX%*b^@{)=-Sa|%n{lbXhF3j=FbtDF^Ahhb9fK{^1inoQvVF04dQw9IX#$028JSJ
zZLk?WbnWOfRS503^`TBinYTjd$8<HYjSB@JS1&MvO$Q+KK%9&|X#ms3z;Fbv2W#{o
zx2{0L>j-e!7$t=u4HJ`SP^AfqK}chMSPeoSq(-<%-V|yOl3UON0)3bfp?kJDR5y|X
z&~>8^G9h%|w!}K*gl+)(Fb2YaUDik`4l$^KZUFiq0>Xe=8zcjuLkj4+(R=U+-FCJ}
zzK3^}(Dfsy22dXm0c`9sA`xy1)@VfUrXh@a0y7G^`Hs&epjJNuTyaEk39=zr9fIC4
zN0?OY1T_h{nU3xd^mZUZySEEeJ966qT|07!qWZ($4aFbGhG6vvdaaLe$TOHpX}}>-
kcpCtpOHkw1*aO8SD5gN2!U`^kq!^?ax`3?PKn)BG0RNTurT_o{

diff --git a/specs/06-Decision-Records/ADR-022-Retrieval-Augmented-Generation/LCBP3-RAG-Implementation-Guide-v1.0.0.md b/specs/06-Decision-Records/ADR-022-Retrieval-Augmented-Generation/LCBP3-RAG-Implementation-Guide-v1.0.0.md
new file mode 100644
index 0000000..14f836a
--- /dev/null
+++ b/specs/06-Decision-Records/ADR-022-Retrieval-Augmented-Generation/LCBP3-RAG-Implementation-Guide-v1.0.0.md
@@ -0,0 +1,544 @@
+
+**LCBP3 DMS**
+
+**RAG Implementation Guide**
+
+Retrieval-Augmented Generation for Document Management System
+
+| Version | 1.0.0 |
+| :---- | :---- |
+| **Project** | Laem Chabang Basin Phase 3 |
+| **Stack** | NestJS \+ Next.js \+ MariaDB \+ Qdrant |
+| **LLM** | Typhoon API \+ Ollama (nomic-embed-text) |
+
+# **1\. ภาพรวม Architecture**
+
+ระบบ RAG ของ LCBP3 DMS ออกแบบให้ทำงานแบบ Hybrid โดยแยกหน้าที่ชัดเจนระหว่าง embedding (local) และ generation (cloud Thai LLM) เพื่อให้ได้ทั้งความเป็นส่วนตัวของข้อมูลในส่วน embedding และคุณภาพภาษาไทยที่ดีในส่วนการตอบคำถาม
+
+## **1.1 Hybrid Architecture**
+
+| Flow: PDF/DOCX Upload  →  OCR (EasyOCR/Tesseract)  →  Text Chunking  →  Embedding (Ollama: nomic-embed-text)  →  Qdrant Vector Store RAG Query  →  Embed Question  →  Qdrant Similarity Search  →  Build Prompt  →  Typhoon API  →  Thai Answer |
+| :---- |
+
+| Component | เครื่องมือ | ที่อยู่ | หมายเหตุ |
+| :---- | :---- | :---- | :---- |
+| Document Metadata | MariaDB | เดิม | ไม่เปลี่ยน |
+| Vector Store | Qdrant | Docker container | เพิ่มใหม่ |
+| Embedding Model | nomic-embed-text | Ollama local | 768 dimensions |
+| LLM / Generation | Typhoon API | api.opentyphoon.ai | Thai-first, OpenAI-compatible |
+| OCR | EasyOCR / Tesseract | Docker microservice | รองรับภาษาไทย |
+| Async Queue | BullMQ \+ Redis | Docker container | async ingestion |
+
+# **2\. Chunking Strategy ตาม Document Type**
+
+แต่ละประเภทเอกสารใน LCBP3 มีโครงสร้างต่างกัน จึงไม่ควรใช้ fixed-size chunking เดียวกันทั้งหมด
+
+| Document Type | Strategy | Chunk Size | Overlap |
+| :---- | :---- | :---- | :---- |
+| CORR, MOM | Paragraph-based | \~500 tokens | 50 tokens |
+| RFI, NCR | Section-based (Q\&A) | \~300 tokens | 30 tokens |
+| DRAW, SUB | Metadata-heavy | \~200 tokens | 0 tokens |
+| CONTRACT, INVOICE | Table-aware | \~400 tokens | 40 tokens |
+| RPT | Sliding window | \~600 tokens | 100 tokens |
+| TRANS | Header \+ body split | \~350 tokens | 35 tokens |
+
+| หมายเหตุ nomic-embed-text รองรับ input สูงสุด 8192 tokens แต่ chunk ที่เล็กกว่าให้ precision ดีกว่าในงาน retrieval สำหรับเอกสารที่มีตาราง (CONTRACT, INVOICE) ให้ extract ตารางแยกก่อน แล้ว serialize เป็น text |
+| :---- |
+
+# **3\. ขั้นตอนการ Implement โดยละเอียด**
+
+| 1 | ติดตั้ง Qdrant และ Redis ผ่าน Docker Compose เพิ่ม services ใน docker-compose.yml ที่มีอยู่แล้ว |
+| :---: | :---- |
+
+**เพิ่มใน docker-compose.yml:**
+
+  qdrant:
+
+    image: qdrant/qdrant:latest
+
+    container\_name: lcbp3-qdrant
+
+    ports:
+
+      \- "6333:6333"
+
+    volumes:
+
+      \- qdrant\_data:/qdrant/storage
+
+    restart: unless-stopped
+
+  redis:
+
+    image: redis:7-alpine
+
+    container\_name: lcbp3-redis
+
+    ports:
+
+      \- "6379:6379"
+
+    volumes:
+
+      \- redis\_data:/data
+
+    restart: unless-stopped
+
+ทดสอบว่า Qdrant ทำงาน: เปิด browser ไปที่ http://localhost:6333/dashboard
+
+| 2 | เพิ่ม Table document\_chunks ใน MariaDB เก็บ reference ระหว่าง MariaDB กับ Qdrant point ID |
+| :---: | :---- |
+
+CREATE TABLE document\_chunks (
+
+  id          CHAR(36) PRIMARY KEY,   \-- UUID \= Qdrant point ID
+
+  document\_id CHAR(36) NOT NULL,
+
+  chunk\_index INT NOT NULL,
+
+  content     TEXT NOT NULL,
+
+  created\_at  DATETIME DEFAULT CURRENT\_TIMESTAMP,
+
+  FOREIGN KEY (document\_id)
+
+    REFERENCES documents(id) ON DELETE CASCADE
+
+);
+
+| 3 | ติดตั้ง Dependencies ใน NestJS ติดตั้ง packages ที่จำเป็นทั้งหมด |
+| :---: | :---- |
+
+\# ใน packages/api (NestJS)
+
+pnpm add @qdrant/js-client-rest
+
+pnpm add openai
+
+pnpm add bullmq
+
+pnpm add @nestjs/bull bull
+
+pnpm add uuid
+
+pnpm add \-D @types/uuid
+
+| 4 | สร้าง RAG Module Structure จัดโครงสร้างไฟล์ใน NestJS |
+| :---: | :---- |
+
+**โครงสร้างไฟล์ที่แนะนำ:**
+
+src/rag/
+
+  rag.module.ts          \<- register all providers
+
+  rag.controller.ts      \<- POST /api/rag/query
+
+  rag.service.ts         \<- orchestrate pipeline
+
+  embedding.service.ts   \<- call Ollama nomic-embed-text
+
+  qdrant.service.ts      \<- Qdrant CRUD operations
+
+  chunker.service.ts     \<- smart chunking per doc type
+
+  llm.service.ts         \<- call Typhoon API
+
+  ingestion.processor.ts \<- BullMQ worker
+
+| 5 | สร้าง EmbeddingService เรียก Ollama nomic-embed-text สำหรับแปลงข้อความเป็น vector |
+| :---: | :---- |
+
+// embedding.service.ts
+
+@Injectable()
+
+export class EmbeddingService {
+
+  private readonly OLLAMA\_URL \= process.env.OLLAMA\_URL
+
+                             ?? "http://localhost:11434";
+
+  async embed(text: string): Promise\<number\[\]\> {
+
+    const res \= await fetch(\`${this.OLLAMA\_URL}/api/embeddings\`, {
+
+      method: "POST",
+
+      headers: { "Content-Type": "application/json" },
+
+      body: JSON.stringify({
+
+        model: "nomic-embed-text",
+
+        prompt: text,
+
+      }),
+
+    });
+
+    if (\!res.ok) throw new Error(\`Embedding failed: ${res.status}\`);
+
+    const { embedding } \= await res.json();
+
+    return embedding;
+
+  }
+
+}
+
+| 6 | สร้าง QdrantService จัดการ vector store สำหรับ upsert, search, delete |
+| :---: | :---- |
+
+// qdrant.service.ts
+
+@Injectable()
+
+export class QdrantService implements OnModuleInit {
+
+  private client: QdrantClient;
+
+  private readonly COLLECTION \= "lcbp3\_docs";
+
+  async onModuleInit() {
+
+    this.client \= new QdrantClient({
+
+      url: process.env.QDRANT\_URL ?? "http://localhost:6333",
+
+    });
+
+    await this.ensureCollection();
+
+  }
+
+  private async ensureCollection() {
+
+    const { collections } \= await this.client.getCollections();
+
+    const exists \= collections.some(c \=\> c.name \=== this.COLLECTION);
+
+    if (\!exists) {
+
+      await this.client.createCollection(this.COLLECTION, {
+
+        vectors: { size: 768, distance: "Cosine" },
+
+      });
+
+    }
+
+  }
+
+  async upsert(id: string, vector: number\[\], payload: Record\<string, any\>) {
+
+    await this.client.upsert(this.COLLECTION, {
+
+      points: \[{ id, vector, payload }\],
+
+    });
+
+  }
+
+  async search(vector: number\[\], topK \= 5, filter?: Record\<string, any\>) {
+
+    return this.client.search(this.COLLECTION, {
+
+      vector,
+
+      limit: topK,
+
+      filter: filter ? {
+
+        must: Object.entries(filter).map((\[key, value\]) \=\> ({
+
+          key, match: { value },
+
+        })),
+
+      } : undefined,
+
+      with\_payload: true,
+
+    });
+
+  }
+
+  async deleteByDocumentId(documentId: string) {
+
+    await this.client.delete(this.COLLECTION, {
+
+      filter: { must: \[{ key: "document\_id", match: { value: documentId } }\] },
+
+    });
+
+  }
+
+}
+
+| 7 | สร้าง LlmService (Typhoon API) เรียก Typhoon สำหรับ generate คำตอบ |
+| :---: | :---- |
+
+// llm.service.ts
+
+import OpenAI from "openai";
+
+@Injectable()
+
+export class LlmService {
+
+  private client: OpenAI;
+
+  constructor() {
+
+    this.client \= new OpenAI({
+
+      apiKey: process.env.TYPHOON\_API\_KEY,
+
+      baseURL: "https://api.opentyphoon.ai/v1",
+
+    });
+
+  }
+
+  async generate(prompt: string, system?: string): Promise\<string\> {
+
+    const response \= await this.client.chat.completions.create({
+
+      model: "typhoon-v2.1-12b-instruct",
+
+      messages: \[
+
+        { role: "system", content: system ?? LCBP3\_SYSTEM\_PROMPT },
+
+        { role: "user", content: prompt },
+
+      \],
+
+      max\_tokens: 1024,
+
+      temperature: 0.3,
+
+      top\_p: 0.95,
+
+      repetition\_penalty: 1.05,
+
+    });
+
+    return response.choices\[0\].message.content ?? "";
+
+  }
+
+}
+
+**System Prompt สำหรับ LCBP3:**
+
+const LCBP3\_SYSTEM\_PROMPT \= \`
+
+You are a document assistant for LCBP3 (Laem Chabang Basin Phase 3),
+
+a large-scale construction project DMS.
+
+\- Answer in Thai if the question is in Thai
+
+\- Answer in English if the question is in English
+
+\- Always reference document numbers (e.g., CORR-LCBP3-2024-001)
+
+\- Be concise and factual. Do not speculate beyond the provided context.
+
+\- If context is insufficient, say so clearly.
+
+\`;
+
+| 8 | สร้าง ChunkerService แบ่ง text ตาม strategy ของแต่ละ document type |
+| :---: | :---- |
+
+// chunker.service.ts
+
+@Injectable()
+
+export class ChunkerService {
+
+  chunk(text: string, docType: string): { text: string }\[\] {
+
+    const strategy \= this.getStrategy(docType);
+
+    return strategy(text);
+
+  }
+
+  private getStrategy(docType: string) {
+
+    const map: Record\<string, (t: string) \=\> { text: string }\[\]\> \= {
+
+      CORR: (t) \=\> this.paragraphChunk(t, 500, 50),
+
+      MOM:  (t) \=\> this.paragraphChunk(t, 500, 50),
+
+      RFI:  (t) \=\> this.sectionChunk(t, 300, 30),
+
+      NCR:  (t) \=\> this.sectionChunk(t, 300, 30),
+
+      RPT:  (t) \=\> this.slidingWindow(t, 600, 100),
+
+      CONTRACT: (t) \=\> this.tableAware(t, 400, 40),
+
+      INVOICE:  (t) \=\> this.tableAware(t, 400, 40),
+
+    };
+
+    return map\[docType\] ?? ((t) \=\> this.slidingWindow(t, 400, 50));
+
+  }
+
+  // ... implement paragraphChunk, sectionChunk, slidingWindow, tableAware
+
+}
+
+| 9 | สร้าง Ingestion Pipeline เชื่อม upload event กับ vector indexing ผ่าน BullMQ queue |
+| :---: | :---- |
+
+// rag.service.ts — triggerIngestion
+
+async ingestDocument(documentId: string) {
+
+  const doc \= await this.documentsService.findOne(documentId);
+
+  const rawText \= await this.extractText(doc.filePath, doc.docType);
+
+  const chunks \= this.chunker.chunk(rawText, doc.docType);
+
+  // ลบ chunks เก่าก่อน (กรณี re-ingest)
+
+  await this.qdrant.deleteByDocumentId(documentId);
+
+  await this.chunkRepo.delete({ documentId });
+
+  for (const \[i, chunk\] of chunks.entries()) {
+
+    const chunkId \= uuidv4();
+
+    const embedding \= await this.embedding.embed(chunk.text);
+
+    await this.qdrant.upsert(chunkId, embedding, {
+
+      document\_id: documentId,
+
+      doc\_type: doc.docType,
+
+      doc\_number: doc.docNumber,
+
+      revision: doc.revision,
+
+      project\_code: "LCBP3",
+
+      chunk\_index: i,
+
+    });
+
+    await this.chunkRepo.save({
+
+      id: chunkId, documentId,
+
+      chunkIndex: i, content: chunk.text,
+
+    });
+
+  }
+
+}
+
+**เรียก triggerIngestion() หลัง upload สำเร็จ:**
+
+// documents.service.ts — หลัง save document
+
+await this.ragQueue.add("ingest", { documentId: doc.id });
+
+| 10 | สร้าง RAG Query API Endpoint สำหรับ frontend เรียกใช้ |
+| :---: | :---- |
+
+// rag.service.ts — query
+
+async query(question: string, filter?: { doc\_type?: string }) {
+
+  // 1\. Embed คำถาม
+
+  const qVector \= await this.embedding.embed(question);
+
+  // 2\. Retrieve top-5 chunks จาก Qdrant
+
+  const hits \= await this.qdrant.search(qVector, 5, filter);
+
+  // 3\. Build context
+
+  const context \= hits.map(h \=\>
+
+    \`\[${h.payload.doc\_type} \- ${h.payload.doc\_number}\]\\n${h.payload.content}\`
+
+  ).join("\\n\\n---\\n\\n");
+
+  // 4\. Generate คำตอบผ่าน Typhoon
+
+  const prompt \= \`Context:\\n${context}\\n\\nQuestion: ${question}\`;
+
+  return this.llm.generate(prompt);
+
+}
+
+// rag.controller.ts
+
+@Post("query")
+
+async query(@Body() dto: { question: string; doc\_type?: string }) {
+
+  return this.ragService.query(dto.question, { doc\_type: dto.doc\_type });
+
+}
+
+# **4\. Environment Variables**
+
+เพิ่มใน .env ของ NestJS:
+
+\# Typhoon API
+
+TYPHOON\_API\_KEY=your\_typhoon\_api\_key\_here
+
+\# Ollama (local)
+
+OLLAMA\_URL=http://localhost:11434
+
+\# Qdrant
+
+QDRANT\_URL=http://localhost:6333
+
+\# Redis (BullMQ)
+
+REDIS\_HOST=localhost
+
+REDIS\_PORT=6379
+
+# **5\. ลำดับการ Rollout**
+
+แนะนำให้ implement เป็น phase เพื่อลดความเสี่ยง:
+
+| Phase | สิ่งที่ทำ | ผลลัพธ์ |
+| :---- | :---- | :---- |
+| Phase 1(1-2 วัน) | ติดตั้ง Qdrant \+ Redis, สร้าง DB table, ติดตั้ง packages | Infrastructure พร้อม |
+| Phase 2(2-3 วัน) | สร้าง EmbeddingService \+ QdrantService \+ LlmService, ทดสอบแต่ละ service แยกกัน | Core services ทำงาน |
+| Phase 3(2-3 วัน) | สร้าง ChunkerService \+ Ingestion Pipeline เริ่มจาก CORR และ RFI ก่อน | Indexing ทำงานอัตโนมัติ |
+| Phase 4(1-2 วัน) | สร้าง RAG Query API \+ เชื่อมกับ NestJS route | Query API พร้อม |
+| Phase 5(2-3 วัน) | สร้าง UI ใน Next.js: Search bar \+ Answer panel พร้อม source citation | ใช้งานได้จริง |
+
+# **6\. ข้อควรพิจารณาสำคัญ**
+
+| ⚠️  ความปลอดภัยของข้อมูล (สำคัญมาก) เนื้อหาของเอกสารจะถูกส่งไปยัง Typhoon API (cloud) เพื่อ generate คำตอบ แนะนำให้ตรวจสอบกับทีม PM / Security ว่าเอกสารชั้น Confidential สามารถส่งออกนอกได้หรือไม่ ทางเลือก: ใช้ Ollama local LLM แทน Typhoon สำหรับเอกสาร Confidential |
+| :---- |
+
+| ℹ️  Rate Limit ของ Typhoon Free Tier อยู่ที่ 5 req/s และ 200 req/m — เพียงพอสำหรับ internal DMS หากมีผู้ใช้หลายคน query พร้อมกัน อาจต้องพิจารณา upgrade เป็น Together.ai plan |
+| :---- |
+
+| ℹ️  Embedding ยังต้องใช้ Ollama Typhoon API ยังไม่มี embedding endpoint nomic-embed-text รันบน Ollama local ทำให้ข้อมูลที่ embed ไม่ออกนอกเครือข่าย |
+| :---- |
+
diff --git a/specs/06-Decision-Records/ADR-022-/LCBP3 RAG Implementation Guide v1.1.0.md b/specs/06-Decision-Records/ADR-022-Retrieval-Augmented-Generation/LCBP3-RAG-Implementation-Guide-v1.1.0.md
similarity index 100%
rename from specs/06-Decision-Records/ADR-022-/LCBP3 RAG Implementation Guide v1.1.0.md
rename to specs/06-Decision-Records/ADR-022-Retrieval-Augmented-Generation/LCBP3-RAG-Implementation-Guide-v1.1.0.md
diff --git a/specs/06-Decision-Records/ADR-022-/LCBP3 RAG Implementation Guide v1.1.1.md b/specs/06-Decision-Records/ADR-022-Retrieval-Augmented-Generation/LCBP3-RAG-Implementation-Guide-v1.1.1.md
similarity index 100%
rename from specs/06-Decision-Records/ADR-022-/LCBP3 RAG Implementation Guide v1.1.1.md
rename to specs/06-Decision-Records/ADR-022-Retrieval-Augmented-Generation/LCBP3-RAG-Implementation-Guide-v1.1.1.md