From bda05feab14ed67b72b42507020630c87ea676d9 Mon Sep 17 00:00:00 2001
From: Connor Rhodes <connor@rhodes.contact>
Date: Sun, 17 May 2026 18:59:28 -0500
Subject: [PATCH] u

---
 _skill-index.md                 |   7 +++
 pov-doc/assets/verkada-logo.png | Bin 8138 -> 0 bytes
 youtube-transcript/SKILL.md     |  90 ++++++++++++++++++++++++++++++++
 3 files changed, 97 insertions(+)
 delete mode 100644 pov-doc/assets/verkada-logo.png
 create mode 100644 youtube-transcript/SKILL.md

diff --git a/_skill-index.md b/_skill-index.md
index d608762..7a889eb 100644
--- a/_skill-index.md
+++ b/_skill-index.md
@@ -41,6 +41,7 @@ description: Master index of all skills in your robot assistant system. Your ass
 | "add this to my brag sheet," "log this kudos," "save this feedback," "add to brag sheet," "log a win" | **brag-sheet** |
 | "support punt," "forward to support," "send to support," "punt this to support," "hand off to support" | **send-to-support** |
 | "add a contact," "look up a contact," "find someone's number," "update a contact," "delete a contact," "list my contacts," "contacts" | **contacts** |
+| "transcribe this video," "get the subtitles," "what does this video say," "summarize this YouTube video," "YouTube transcript" | **youtube-transcript** |
 
 ---
 
@@ -214,6 +215,12 @@ description: Master index of all skills in your robot assistant system. Your ass
 **File:** `skills/contacts/SKILL.md`
 **Dependencies:** `uv` CLI, Python 3.12+, `contacts` script at `skills/contacts/scripts/contacts`
 
+### YouTube Transcript
+**Purpose:** Download and summarize transcripts from YouTube videos using yt-dlp. Cleans VTT subtitles into readable plain text and summarizes or analyzes content as requested.
+**Triggers:** "transcribe this video," "get the subtitles," "what does this video say," "summarize this YouTube video," "YouTube transcript"
+**File:** `skills/youtube-transcript/SKILL.md`
+**Dependencies:** `yt-dlp` CLI, Python 3
+
 ---
 
 ## Adding New Skills
diff --git a/pov-doc/assets/verkada-logo.png b/pov-doc/assets/verkada-logo.png
deleted file mode 100644
index 570ba79fd2c51454afb0627324945fa0a7612af3..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 8138
zcmeAS@N?(olHy`uVBq!ia0y~yV60(aV3g%xVqjos(DhDYV342a>EaktG3V{w+yIgC
zdF&tVt)6>p;iQ#YcBbBoky>+e!!k2#=7=9oEaw?ir!t#v6*7->{jz9|p_(?wN5w^f
zJ^X@>iw$LFxw_~_%5G~im@0Kjad*?R){rTQo40b_)J)uzpH`N-rmWlZEBAVj`-L)>
z|IVMs{_isX{qw6^BwuVxe`dY^fAaqa7Z<zpFfi!;cX{i<z`!BPuz+_eA45Rv1xAKd
z0#*zTqE74#E1X_1Foc5SxE3-qgeaCUG-$O9K}CNjynConXLG*(>+|&zZ70v1@z@?!
zs9ET1!nJTQZ*AW9iO%nT{H<$CjF_eyy)o(iKDNmxk4Qv3F<n```Cih-CBOD4bjjOP
zebKm8tZ;PKoO6G9c^{tqa7KRG)amk8MK?Cg-)OgeciHc<7-3zf`F(1_?THfM`S0)D
zt*)4t-n0Dl%N`$@=6GRV$EAx;9+?>y7I*K;rJNOI(^7hqKV&b`T~T5($Kp|Wb*%5&
z!hN@&X2~eW3!imdy7=J{!&~LmSEP<NTzBaXc)22Q`sJOQ*Tp?tep=!FTdypc#j(P@
zhbHXw%AONZfBNd}(*`|p`)?<&`pr}1yM_C#w&^vC!vDMe9lOZt_+!m}`(;w@-><v8
z4tSZd@7Trf<-fPSduNd)QyVY5)_H#4wrGK0?@qrc-d*VXg?-`T)kkJpy88*R9P_X}
zo-)IHh1`qF7o#7&SgfzS>|$YajPPBDrHjuWx%j2Lx+q(wHb&Ui5zK!lv(s`_VVcw4
z$x$mxzRa<h^*VgsZmD#&E7mJpz1+`tE?!&J?$o<@XZ9lR6(uGcP95a<$~RZaJ^H%K
z`hb@&_8hx-q}=+cl)LbCm*Zf5;Qw;#zf$ghue;pez~kh-T1II3Hd98;HNQiueM|VK
z_BGnYX8q3dzox}*8!N2r@Ipen@XBUCrLAjMc}Tb~XOlhI<}G7-W}<iEc|*x(yg?0K
z#}{-SHt0ECY^$2hbS3QZinVN4Y(uPyH6%D}_U|zf;WE^6`?)*ztftM|h|3?Zii)nZ
z<G$;()VMwG&W?qjjx3CMpP^JJ(f0D?zG<(5FBtR8JioB>_jk#w-2QDZdn8w<uhCk?
z&M&v*m&g~Mspn))os2z^aU`Z+a`jGm^-GWb{0W)gG*7YQSwNXzS!AW<V!`6;?W?AS
zP3=g%*=4G~?3d0Lo`uHkK5Ca2Dz|fUaZ55^x1Qxa=kW2qTRGc8=6B3X{E~I<qs{+c
z*RHd%vq|)NJd*OrRqeApd#C2plTVi}h0O0br&yA4A?clpMc=$1e^&(sJys~xaQNKC
z)_t_-*PGwBveH#=>@QxsJKDK3+0R(HaE0TGfJHm+e5_b^{Q7=nj;1!Z<42{~*J{lY
zu=<jrU4G=Jb>SIlmtN+S7nc|6N^o3xzv8mj|31sJF)=ZVHpE93|Fw8w>BkvjwPk7a
z+F7g1*OivP-gL}ywU(-Ym1e@mh`#w5m%QGsS-U1?XIAKy<(aGH**hmYam>AP!E5j4
zZJUbT?`vC;tDU>H$l-ip)<S>vsUi3FFE%I)*{wWX<%)%7yO;XZD>3iatX*Ssp+dlv
zPeoR1?v>z&HK|uD49WvuE}3`h+4~npg%|o4%W=qt2Kc{wvFW+QYMBt$g^R_qB;3Q}
z9%p2f$e8xsTHx;<8+2oT@ZMDKvwcyjrhRQf?8%{8Ene=iL7^b?Lnc|QC{!q1QF5iD
zXz$|FQtZDUuMpFcU3}fTYHi^L`@XD&mkdkxHcPR;aDE|CKDXu2k&D-AWWLC(mY)^Q
z7ONh(H}&-F-DR(G?D*V{n=(%A6TBU{D<^c}v5Q<SeH)98&q<4m{rITyShLk_W&zvd
zQccgMZoKBXX8o-V<?+q`zA+q$nA)aL!ei36ZqM@w=e1g29%t~)3U|v)UntKW!n82C
zqp(D^v$;-Y0rOeyzx|7~*A#BK+TT1c`K{B^#h<I#P5D;vYRTGOw_Z``&>Haa#kP9~
zwb%pHL#%$SDc^94F@$^J;=@0!lX``}if`R|iqZD?uEu9m?>&3pVd{6~d3N?&3n6!Z
z0V^v-4krtlcg4HI*F`PeJavwT%v{G85`E9I=3Fa&{z>fV(qAqLnHQEP>&&_>&)&J%
zenklniz(j<MlIRJ4!z7;jb6tE0v<P1Ex2^i{))u{Ee_ea!Ym7y|DM%W)yFSy-L+Ax
z*7>v7|E5NVuglIV$LTN5lUQxU5n}agdfwJ;bM4d1EhI#pA4#w~9e3F2{M_znp~E8y
zu36!1i_Vpa2_4S}cq!4r^;%Y?k8$B8Lz%BvtUY9sot7qBXss-tVA0m<b$mfk^m~QE
z9-gUvj!O$<WIK}!IAkAtZI9BA+vl=#ZtD?7ll47)Y1Z7G%qvP7;`TNlVHDJoT@d(Q
zp)f(}S4qOT*Rm>nXIi}2pKZJ6_~?Mnijo9AkdTYh3x_GcTaWQ}GEWk)GKk)FbymeY
zg+d3uUnL5sYP%lqaPW{(a9GOhnU}6u*uXQDPi6Bx$43q!D@qhHzw!yOPjPtRAo4Qo
z^~=rg-OtMQJubLk7{GK^J3s}Tn<i(cFFeNhRKRLN#(zE`_N1m~q8h0iBb07OddPH$
zYRPu=DsafYeY-AB{IEgan>VX^rE|=r^^`Q)K{fP+Jxqh3nts*ZW5&wdXuSOV3EQZx
zh0SNvyni{?+Qqy7X5Hz&eeXWqyIH9wtIxmPXIWwV>AB4JE6ZNJ3Yjyps?>7sy~O+{
z=L^?<k+FKV#pL~m4;|r;TGJ(-Z9FblU3~rA68S5YPhUMf=XbK*qVM+Z^^F%d&2BbV
z{a;$v&b-jLyZnZB<ytN-?$u^_CNq7Czbd_K3Y#4+*1dGi9-ns?{bTj!J)d0dGVi&4
z>CPu|vC%vE@=Y0R+xZR~xT>l?Guiyaxo-de&(Yf_Rqr^TfA`bxzmxc9UOwZhs@ikR
z_S=o*m3FKKEWGV~XVXrfj#l};V)b(Ko$bnh4({3GDs$XW>U)3vALXmd8QK;+ypSl7
zc68O$BK}I7>&=Z0oA^X{SZwCKZn^Amo9lqU!p(cOOi7JA*?ws9!*~96p4S@l?(7hp
z$K!2xWYe2ZmxA1WU)aNBwMEt{`0h#VuoJegUX^k-P52SW;<%$Eu<Alr;*3qA+B#y~
zi)%CD7-etq&0V?d?5U|mcTX-}9M6AW*6)wN;qrEO_x);f93HhZ81S8KKi8BlmXi86
z<kP|rDLd{z_WUF8SFS(i_NjLTIu&V*i@8DO%;KNBLbyyQCvoEP=ToLmf4Mlf^_ghE
z;-4VDT;<jmP<vBz?f(1|^4I@<kzqLXMZoH1*y}sKHx&KP&+HbBTiNtX^uogS3yBiX
z!e3uIWm{5QaN+sqqu<k;8xNd4c+i4xm8-;FCaV%5tDQS{78N%+FAVr!db6@N#XRp0
zN2}8${*5}jE$r?4_MiM*V<5o6#^fn^uD`#6Pt%6+DU0KS{qCO&R_-!#^6K~N4GW)n
zep236z1VF&d7L{|Ul%=YmFXx@wQIeqoo)N`pXc-M?y7$NF68@#9~D1#|Ci=Ced<@H
zo7=HVssCr~E>JvPxX~|L?Ej`G&fT9sr_Y<Gz;U>|{kQGBna>LzCZ*q7v`58keN*Fs
zxz^<z{uS$1XuR4VD8RD+e*N=q{k%W_`0qV@_~3riqw{C&jfy3M-|$rjdV0!C?=IN!
z@AY+dcDCd1?#?dVarEa;*P}<d=S3R0A7+KLE{|MSH8C`-{qjYAdHT8e|2L=4KYQ}*
z%&$SO8vLBx=jbboTPeq_KmWXY>;I$gzw`6nomO*VPFk3e?eXWIE!W5Vz4CO@>n^#E
zRqlJ@!t>W#o}Dv&`-||`*R*~{_;Ir}Z@s<Uo9}sj&DU2Y3x3q>3qGqWcfNFI<NqZl
z>nrB1zaMCBVW9AHh2`0r1se@t+2_9ac46lCd(u+7WGh!aJXxS2BBUSw`X>K05s{Fv
z(6c;eEi^1+qO(`ueHgKQU(wT5p`I~s4t@TUZ~yB>(R0<4-`oCIKRUX_S|;^r>i(La
zsdrua>zvf{1Lxg+IN^5U@ye%JYpmZ`tdE)cdrDgBUlVKf*&lY>oUi}z@20kVz0P&3
z*{U{@axxMUG7=gZI(Nz+yjA(;Uv4?qop0~r4<?U}eg1RrZuJwP_QV-)O-)y8Yn^y{
z?BL&g@4A1=txk&J`R~sjJh(|z`_;WQ{f{qe<S|`;I;FNVUFu`icXN|3CGpmEi!?vh
z-l~}QzIgZfIaU7eR%?8Ep&a-A={fJ?vu6wQL{7cnIloV%)A|iZlf%4u&Oa_m@3P(U
z>Heu(QPGcotl;P4S)-%<y2xr~3ujZozO}jOv3n0)?$}+a%Hec*_wm@hn-14pjoucL
zd&lF)*S>GxrlzJgGc$WBB}8e6?7DMvXWiRf5$s$wcJW`Aot<uNJyqWCPgum}T}dy#
z1^68Avib5lI{tQK_<DoFWk3J^<9)4te*4y!3y!;oFANR|QriB{_?m2)-Rt-2dvwG0
zerUSm9P9Ud+S%F3wcck%=JC8=v%3EKi;XPz<7D5yb?284{P$<a>~LN6Sx=?5-DfYU
zJlhwwvqy|&;ksqtUOZGbJbKn=@#6UXL0cVv_A%Mbe?N5<Z;`0s%=_=xuV33+ZBX*!
z+Xd}VX^yW7krhTYUp`Fi`&Q=mxJ1;-Z1&sdWz|7yZv^JW1Vm_rCzo5!J*CaPY1_4H
z(}d3!{;%49TsZjAk+(I`(b3tf?<`Jz`n0`IW}3Imyyvg8vN%;I2|YZy;r{#e_qIfO
zs!H>)#I3wF;Y4`T4IY`@<@(p+Wp>$4d9Z!rix&@-JvCc`J;E0{&$E8~F(&-RAtnFw
z6K@|`e*W#f`xDOOPSP)n>G=KDm7__;Z2hh|=O*jw7WLnf?Q?eLJ6*cNC3nivl#(y6
z`S00#*YzoIe6@Zh^#4Zp-h9KqUFV;#Ucc5i?EI~n!S9d8=B{kK=agPpXP4MM_kP`%
zz^2Q^7JbpXN;03EueG~BZ|?3z+XT;KroF2!Qn!5F(l1lhe(SdOwP#Z$Hnl9>ko)|+
z;I4|p4>wGzYWy4jWO=LlVR37{3F|}MtE=z$&St4Kwz9I7@vF<<prgHgS^BK|@9nSN
zS#<ZD?BAZavv(5P4HDf?ru}WXTRcrKTI}XFA(n>^{%z;HpZv_ExUkUkX4j|Xn>Xq7
z+w1)+$jSNhZm#|lHec6;0WU&|y@S1rgV?ey`%cfX|6kuHSIwNbY~gL^mI<km-;SIV
z`{dAnG|94V(aNgJX(!)__C21budS`6ckaxY8Ce~+H5LEgf3IE_Tm1FQiP+~KQg$r<
zd#LeP^Q3+@=EiHA<z4=k-<0h;`tzsdPRmFC`}%u%`FPaiLQi{p^P8BxTeEt4&6Wil
z3XZOtx|3VH>pkzkhlj1>Yc5`{?JbDuf4=Mf`|B^XWcIf5RUYcwahR|2&c)}SkM64G
z`>7q0CsLDj^2VY)t9qraZ{7Caw&g@^L+swYoPHjOk!5GS^+VTGPPAF*SNZwW^Ut|A
z_sld1-~IKDrRx0llGR(^{@EG1^KjvvZSngJvx9tgUA5Y(`fbsk-1Yg&XV{9qemQZs
z_RX1>#!s&KIwn5cW`3_@&$-#FR*U+TC<&+YyqO;;-4*Y;f8E-drE8DWzd04U+V1uH
z`<siWKgn!YKU{c*DUfroDX52k?&LOWnZoxQoM+5CTp06Y_N3=OPtTqH<X==|<<FO!
z-wWF{s!d**_4pWbZ*R+@U7Hs-cTT_iFvGU|-Pu^{Tr=sNF+n%>aIrC2Z@Ye?w&85%
zv%|;xwE8{D?(Nz3`T4Tx8>UR19{;jv-MSSMtgn`={`%$ScWy3jRe9ce@o95}&VJsr
zdb#^MnKPNsVynyM|32GZv^;n7%~?^i_)W~TGDU>ftXpv-E?D`W`RPsUQqr@dW<9+p
ztu{HcGV|rbWcR1{Bs}-u=QOqAk)3=}<a^8973SN!{v^)7t$W>0>HCs%HV?0gioRZL
zrcyh(a`)lFkDr#RoYUZDV=CfZ^fL3f;aiDZqnVf1UgNjcyCK`Rb?Q`K-Lrw|_ak+t
z)w184_UBK#`r-2TcY6O+&cE%eQM~Q?m7G=U)YLcqkP=jKSm=B8>h5oEo>{E>ddJdf
zn_qNXl-BoSANu>%KVD-td1+80G;@RV4cqqA$dh-AIbLs$tIb{g>EtF;>#V2O*0iY~
zHqX10kfc(!M#H{w#aZpbh0KX<+?BTs8_%X?Ti55k-^bP(8!mHPS5sp`&5QN->!dGd
zZ`rmf=;y9IPnRyeywvN?Dzm&jm!5|F^9|Ve%=P3F?s{=KY3XzmkNFFeH{X16Z^{!(
zi<b7fyzTB`hYmVC=vP*rbY5RWr)J%Xj=(c1M!CPgN*DFt`|($PhD7eoUs*n~y-%J^
zRp5Bi%J(Gmc;M3LwSmI!F*5cg)hib7t*wdq_4(U_gUykLtqwXkth-*b>i>(cxrTF}
z>&MqcuKOq#TzFg~BJbVjpTF+pJ)gRI(s@q*w)gk{TQ7FBdgCEBb*;wq*~~@71qKov
zF8e;%oEOf^Q?+D%ac-`4QR(mV^J>kmX)4Z~`8jgh)IZISA3ttV(2zfz6!~qs{{Ay6
zoeCVf>a#eDvrb&@6SiD;|NZ~8l9HO7oFmpoc6N4Fzf5<Ue?QpFzPZlG%uMgQlTCb>
z*y|{*?xUOXYM0)z2vKTkJaFU2j~h2^-)|^?b8qj)>hFv8)vtK_V`9uM-nmDA{**cH
zxi4p7z>43K{gt|3*=?U~z^1da`rMBjGwyK4yxCFk^l^W()$<SKi93Zke0^Hf)TY!s
zO%oA0vWsWZg*~q_^3~6nJ+bIBcx#~+yw%5`;(53JzfWi5|9yEn{ZX87XOE;IkFy4Q
zS6BDrAI&?t#ZT3mYp8QLWh6PxF;jfDa>rATrvDMX7jj=j=uCTldRbVkn_Y^ZvAf@m
z&iwCl1lO0`_F7Xn`3GB<)z%!b-Ogo}9&667zIbuH|F#-myY1)m?|OacYYyDb*g5--
zIP>XEPu4O$xshYmeN?FY1xM3|$ol{FvAfn3c<%l8Vza+{NJxy%U1oDhxsO%pdv|i~
zelC!gCmMO?{KO9_GwOc?{QO*8B(C=H!jtoD7R6TYg62Q}S>7)f{ki*U?%K7U{&U`I
z&*eNhzgSYTFg*Qou}-pSNp;5O{rfk4wcmDR*OEB@FEUm)EpHZWKO*$}@tI%ohkE|~
z*eA4Ab>dv+zjgJOcOQTJ!*%}o?R9T={n*CnxzNJN{Z3}e;a{ghbdLt`+qU~%;k3Sd
zSf=yMJZJXPn=Zdrf9BC-%^vse&70!g$-kz$l+6rycT)Yi=RX@0#aKp7!Q=0{tb1E}
zqwY#9@^|0zs#W|+m_*0+BSO_mZSVPn^YUD8*go%6pO>&=Lxj$=EhcVv`(&!r&CSCT
z^S6JBRXdZJ_PPIzSohJ1bN}l5+fJQ4`*TUyw9lumT7+HtzhC%Aw81jQgEOD7y?I(<
zIhU(k%5eG2%k^{4otypq#JOiXmKPO?+x<N6ueL+B{niVcl1po^J$fXxb7Ne%|9T0l
zmQuFL%5IMNO1ls1eLucDjOYC1e|^2qZf<Qi=Uj-nkz%?&V%k$9BbC#_`|sDkNO-)`
zc=^;nbB&jupJ4A%)_!>6+_xvnQqs<yKXWGG`PpuL1tTM4)q1&kJnpd_>Y5T9Kax4;
zCMXm<xnSL*zCmaAk_89eZAec^vzt40>b!Y&#aEmbU+(Lbw(?SY)A3*M{@Jxp?p=5!
zGhvqOR-aF2N`?JBKYPVQO`Sda_n%KeC*vggpI1#U&iW|w^F0f5qu%qYF4LBKowMmv
z;W%_TGP0=uTeVGL3t!@MVb8bwGpASYyZf->>(;eTe%pTFm(MD(>l6H4lIC|~VvU9f
z*WDj-iQi>Don2d0T>SJ}bVs+Y&&k6p4t$$6>*>;^FT-BH+neucr)VP?w&`fdclG%m
z+bWNWz6zYaoU3WUhQ?&0nV<^)JKK*dPrJ7EJ*b?w;dAGc&rf%8b8~U)$L-s)XUi0u
z#jj>x49xwy$$jbn3xeuzPS{qxOj1v*I(I4c^ojhAlGUMk%~u51e~&6R7oI-l&3Tu+
zxS5yF7>ir2iP^ep%^sU~y31nrUYdV+lc@HTDRPI)+c|IdpP7`m^%Sq!&%ghSm!CJ-
z_~Wg+{Lw3)|6ZNHe^QR?tjlM9^X*xi!u0&?!P!0gUf7o$Q7taw@2%L|5^buvbLZ4M
z@0J$kYu|icCU14a!@m5)+0~C8O`6Fq>!Wu0UM+7?=)Be2rwi<`Ie)tMpT#Co-RxI8
z%XF6|Djdo#*||CR=e!NIdFF}LVV=p!mrn0HV{fD#Df#^KPmb9|8B-LU+8^(*5PRvm
zcjwfM%FLB9U*CQIU7oW(aGrIOLW$+vt+%(wzk3sMb=iN`*4ByKvK+lSg(}|f<mKkg
zneuwcj2-p%uTFm1dADV@zem)r63s;KuD`szUU_dXJuT6?IWg?#{*L}@!O9$mRz4G(
zyLD}kkd@lx%)7r%Mbvvg{36qR)G0#N?$OtML0cV9o}E-<7a!~U_S~+jkGZv*ma1OM
z)GxlW=aI#`ADwC^+ZYy36y<8=dsiR*X>yLA_vGv_9zLE)_J5qWx1UgUoOEgJwX34~
zlk(pNg$G-gRC|1SVK+a&{_wsZ?>-y8KdG)-FUQq-VtxJ<*KN0!{=c~ThuF)B%b%aw
zv}end4U5-S%--_o(WFzWR=t_?Cel&uUh1CBiyuFG@Zij4;p<{HbEmrI%E#<{CTg8O
z@6Hm&jT4{BPMi1C#KPdh#d~r)H%v45W^Z9|p=eQLp7=St?a_I^&#hQxmZ$J$y^a~D
z?Y!5Kk@m53=Kr6(ynfP5%m4rW?(Z&N(7$WT+=dI;#m}x(@8lNu+PBZ<y!qa0n`=#e
z+C4&6kNB)l?Kki0_MW2uQtbOCP{G(@dpLIQ-gWolrlr>BZMXk;MEK2~%7vYs?EG?n
z3R7I(Z9j7D=F!vBuha-<Z=9%F?)~-JW_gjRQ_lQ~nd!4^>FH@FYZu*$ivIuW?e?jm
zq1^K#6U=7&ss$Se>?nTTw>te5)3MrlGRN0ezEu5G<5U<JSCsByKX<C?_uS{-e_#KT
z^mW?W^<q<}bbMw$vexF;`gryH>+4J1IAuSz$evU`{^fYC*=<k@$*%gc)4~8B?|ipC
z_m-9SYu^vjSIu5OPg`r!>_$nh*0P-SJ1;$*Q~J-FYns5mV@Dh2&5M<sHN)eNaE8&$
zxZ2#;&$s8D{kn?#q_o4Sy)PC9fchwF)9p7z=sfy<Aw+-oJFCK;8_R{a9yh=C(z2+2
z{+|6+GW+(6UkWZ!dwFE<H?N$Old-o2*PCw&uBpt-yt1x#|9$;yvJ*R%?<r|r`@AJW
zN8NARk&oS*Z^-iPocmBUx7vBzX@3!})+6<6?!WK%k=cEC;vMtQQ<54bM<Ne#%@>RB
zPAheFee|7Y#@6$dO4SWP&T=QSY=1kZg#U`>-K6|3<(bLnZ?mLY^`GjP&3;=^T&$!o
zaC%d|smYNor);d%t8-J&JeE-Ed%WzO!nemIZtgvzRz=k@?smHGoY$;dvE$S3bu8O2
zitfL!7xN~h=rDt6POhSuev$mz0F5tx$&c3vKHB42x%>M*(bwfWEYI%Qm(~1f?G@#x
zx~H;jzt@}Z?d+@&Z;uG~o8hx;m-*{ttv(l-<Fz*m6W5iCFE}5u(75#Rsy(hxp5Kt|
zW6>_?d#o3?Z%g+1cR%(8MozoCHNCtjU)DJ>o4YOX)x9<9CYzMro62P;YTVP-`Y4)P
z+5O4x%?<_g8~VrJ71Z1)RNN<iG-=};yWe#_?m8Mee>OaBa(n;KGi}WU!<jxUCs&@l
zWt*P0xBqC+*22fjYu~-njTUj6VR<&E+@kF5&A*xRtAk6o&T?PlQO=c@H%&WEuPmk`
zC#ADB(J@hR=aa<9vR8IB`(Jy7y;Q$v|Nc$=Hm#MOmVMUOO8$6FH=f9`AwtJj_w4am
zeeSjP1v1;S&*y!zl`?(H-Y;jH)T1K7G2y+UXenddnj<qU{>q)7zwghdgU#%Z{@s4$
zQc-wDNm=>uyA98u==IDw>U`|;o;xg?mwHb>RyDtQ;ce$h#ou(rx<CExKg!$i^x3Vn
zv~$x}i?rqOf33C&{w^ymFK!)tbZ)BAb5LWpL#AAP&f8*F*Hik7D-M~n=U25azgFza
zZ`WmAVX<lRbP=7-Kh2#n+Vh^<?=L-l_3~?3xq0`>;%s8v+}v(<eTwWp$LcsCI#|j4
ztou_QaVulr_YXzBr}=G^)05$AS65$NpXBE)ynK<$>-Vcu46E)eH+d~2_)JfIZbfz5
zO#7La7jp~#4{CkSBD(L?Mx&A!3=Fl$UR`EvnBqJ<rh&1~UrWsTC%tR!_fN-o+x9as
PFfe$!`njxgN@xNAq?xe>

diff --git a/youtube-transcript/SKILL.md b/youtube-transcript/SKILL.md
new file mode 100644
index 0000000..5276d84
--- /dev/null
+++ b/youtube-transcript/SKILL.md
@@ -0,0 +1,90 @@
+---
+name: youtube-transcript
+description: Download and summarize transcripts from YouTube videos using yt-dlp. Use this skill whenever the user provides a YouTube URL and wants the transcript, a summary, or to analyze the content of a video. Also trigger when the user says "transcribe this video", "get the subtitles", "what does this video say", or "summarize this YouTube video".
+---
+
+# YouTube Transcript Download & Summarization
+
+## Overview
+
+This skill downloads auto-generated or manual subtitles from YouTube videos using `yt-dlp`, cleans them into readable plain text, and then summarizes or analyzes the content as requested.
+
+## Prerequisites
+
+- `yt-dlp` must be installed (check with `which yt-dlp`)
+- Python 3 is used for cleaning the VTT output
+
+## Step 1: Download the transcript
+
+Use yt-dlp to fetch subtitles without downloading the video:
+
+```bash
+yt-dlp --write-auto-sub --sub-lang en --skip-download --sub-format vtt \
+  -o "/tmp/opencode/transcript" "YOUTUBE_URL"
+```
+
+Flags explained:
+- `--write-auto-sub`: Download auto-generated subtitles (use `--write-sub` instead if you need only manually uploaded subtitles, or both flags for either)
+- `--sub-lang en`: Prefer English subtitles
+- `--skip-download`: Don't download the video/audio
+- `--sub-format vtt`: Get subtitles in VTT format
+
+Also grab the video title for context:
+
+```bash
+yt-dlp --print title "YOUTUBE_URL"
+```
+
+## Step 2: Clean the VTT to plain text
+
+The raw VTT file contains timestamps, HTML-like tags, and duplicated lines. Clean it with Python:
+
+```python
+import re
+
+with open('/tmp/opencode/transcript.en.vtt', 'r') as f:
+    content = f.read()
+
+# Remove VTT timestamp tags
+content = re.sub(r'<[^>]+>', '', content)
+# Remove timestamp lines
+content = re.sub(r'\d{2}:\d{2}:\d{2}\.\d+ --> \d{2}:\d{2}:\d{2}\.\d+.*', '', content)
+# Remove VTT headers
+content = re.sub(r'WEBVTT.*', '', content)
+content = re.sub(r'Kind:.*', '', content)
+content = re.sub(r'Language:.*', '', content)
+
+# Deduplicate consecutive identical lines (VTT repeats text for overlap)
+lines = content.strip().split('\n')
+clean = []
+prev = ''
+for line in lines:
+    line = line.strip()
+    if line and line != prev:
+        clean.append(line)
+        prev = line
+
+text = ' '.join(clean)
+text = re.sub(r'\s+', ' ', text)
+
+with open('/tmp/opencode/transcript_clean.txt', 'w') as f:
+    f.write(text)
+```
+
+This produces a single clean paragraph of text at `/tmp/opencode/transcript_clean.txt`.
+
+## Step 3: Read and summarize
+
+Read the cleaned transcript. For long transcripts, split into chunks (~4000 words each) to avoid truncation, then read each chunk.
+
+Summarize the content according to the user's request:
+- If they asked for a summary, provide a concise summary organized by topic
+- If they asked about a specific argument or section, find and explain that part
+- If they want the full transcript, present the cleaned text
+
+## Notes
+
+- The subtitle file will be named based on the `-o` flag plus the language suffix, e.g. `/tmp/opencode/transcript.en.vtt`
+- If no English subtitles are available, yt-dlp will error. Try without `--sub-lang en` to see what languages are available.
+- Auto-generated subtitles can have inaccuracies, especially for proper nouns and technical terms. Note this if the user needs precision.
+- For very long videos (>1 hour), the transcript may be very large. Consider splitting into sections and summarizing each before combining.