Ababhalisi abathandekayo, namhlanje, ndifuna ukuthetha malunga nokuba kutheni iimveliso ezininzi kufuneka ziphinde zisetyenziswe ukuze zisetyenziswe ngabathunywa be-AI kuqala, abantu okwesibini. Esi ayisosithuba esilula kum ukuba ndisibhale kuba: Ndichithe ngaphezulu kweshumi leminyaka ndicinga ngokwakha ujongano olukhulu lwabasebenzisi. Kodwa "umsebenzisi" kwi-UX uyakhawuleza ukuba ngummeli endaweni. Loo maqhosha e-pixel-egqibeleleyo, imivalo ye-nav ecacileyo, kunye neeCTA ezibekwe ngononophelo endisazixhalabeleyo malunga nokuba zincinci kakhulu. Makhe ndichaze ngokugubungela: Kutheni ukwakhela ii-arhente kuqala, izakhono, kunye nee-MCPs: I-primer5 ekhawulezayo amanyathelo okwakhela ii-arhente ze-AI kuqalaKutheni isigwebo somntu kunye nokongamela kusebalulekileYintoni onokuyenza kule vekiNdiyazingca ngentsebenziswano neGranolaIintlanganiso zakho ziqulathe owona mxholo ubalulekileyo kwinkampani yakho, kwaye i-MCP yeGranola ivumela nayiphi na iarhente ye-AI ukuba ifikelele kuyo. Qhagamshela nje iMCP kuClaude, ChatGPT, okanye Claude Code kwaye iincoko zakho zifumaneka naphi na apho usebenza khona. Yisebenzise ukwenza i-scaffold features kwinto ekuvunyelwene ngayo kwintlanganiso, ukwenza amatikiti asuka kwi-standups, okanye ukuyila izindululo zemveliso kwiifowuni zokufunyanwa kwabathengi.Zama i-Granola ye-FreeWhy ukwakha ii-arhente kuqalaInyaniso yeyokuba ukusebenzisa i-Claude Code kunye ne-OpenClaw iye yadibanisa ingqondo yam.Endaweni yokuya kwi-Mercury ukujonga imali yam, i-Linear yokulawula i-textile yam yokwenza, i-Claw yam i-Claw kwaye i-Ila yam ivula nje i-to-dos yam Iarhente ye-AI ukufumana olu hlaziyo. Ndiza kuthetha izinto ezinje: "Bandakanya iimali zam zeMercury kwi-imeyile yakho yenyanga." "Dwelisa amatikiti am eLinear kwingcaciso yakho yasekuseni." "Zeziphi ezona zinto ziphambili endizithathayo kwiintlanganiso zam namhlanje?" Abantu bayabondla kwi-AI kwaye babuze imibuzo.Idekhi yakho yesilayidi enamaphepha angama-30? Abantu bafunda isishwankathelo se-AI endaweni.I-UX yakho entle? Iarhente iya kuyitsiba kwaye inxibelelane neearhente zakho ze-APIs.AI kungekudala ziya kusebenzisana nemveliso yakho kuqala, ngaphambi kokuba umntu enze. UPeter Steinberger (umsunguli we-OpenClaw) ngumzekelo omhle we-agent-first thinking. Abanye abaphembeleli be-AI bajonga imbali yakhe ye-GitHub kwaye bathi wakhe iiprojekthi ezingaphaya kwama-40 ezingaphumelelanga ngaphambi kokuba aphumelele nge-OpenClaw: Engalunganga. Wayesakha izixhobo ze-40 ze-agent-first command line edibanisa ukuba i-OpenClaw yenzeke. UPetros wayenokholo lokwakhela ii-arhente kuqala ngaphambi kwakhe nabani na omnye umntu-kwaye eso sisizathu sibalulekileyo sokuba i-OpenClaw ihambe. Ezona mveliso zihamba phambili ziya kuba zezona zilula kwiiarhente zokuzisebenzisa.Ii-APIs, izakhono, kunye nee-MCPs: Zidibana njani phambi kokuba sixoxe ngendlela yokwenza i-arhente yemveliso yakho ibe-friendly, masigubungele umahluko phakathi kwe-APIs, izakhono, kunye ne-MCPs. Cinga ngee-APIs, izakhono, kunye nee-MCPs ezifana neenxalenye zekhitshi lobuchwephesha.I-APIs zizixhobo zasekhitshini - iziphelo ezithile ezivumela ii-arhente zisebenzisane nemveliso yakho. Slack's API, umzekelo, ikuvumela ukuba uthumele imiyalezo, uluhlu lwamatshaneli, kwaye ulawule abasebenzisi. Ngaphandle kwee-APIs, akukho nto yee-arhente zokusebenza.Izakhono ziiresiphi - iifayile ezibhaliweyo ezilayishwe kumxholo we-AI ezixelela ngqo ukuba zenze ntoni kwaye njani. Umzekelo: "Xa umsebenzisi ecela ukuthumela kwi-Slack, fowunela i-Slack API's"chat.postMessage" endpoint ngesitishi kunye nombhalo womyalezo." Ii-MCPs likhitshi elixhotyiswe ngokupheleleyo - badibanisa uqhagamshelo lwe-API, uqinisekiso, kunye neenkcazelo zesixhobo kumncedisi omnye. Umncedisi we-MCP ka-Slack, umzekelo, uveza izixhobo ezifana ne "send_message(channel, text)" kunye ne"list_channels()" ukuze i-AI iyazi kakuhle into enokuyibiza kunye nendlela.Nantsi indlela yokucinga ngezakhono vs. Izakhono zigcina ifestile yakho yomxholo ingqindilili kuba iyimiyalelo ebhaliweyo ecacileyo. Kodwa wena (okanye i-arhente yekhowudi ye-AI) kufuneka ubhale i-auth kunye ne-API call logic ngokwakho.Sebenzisa ii-MCPs xa ufuna iplagi-kunye-udlale lula. Iseva ye-MCP eyakhiwe kwangaphambili inika izixhobo zakho ze-AI ezilungele ukusetyenziswa kwaye iphatha ukuqinisekiswa emva kwemifanekiso. Kodwa ukudibanisa iiseva ezininzi ze-MCP kunokuvuthela ifestile yentsingiselo yakho.Olona cwangciso ludibanisa zombini. I-MCP inika ummeli wakho ukufikelela kwizixhobo zenkonzo. Isakhono siyifundisa eyona ndlela ilungileyo yokusebenzisa ezo zixhobo zokuqhutywa komsebenzi othile.I-MCPs yinto enokwenziwa yi-AI yakho. Izakhono yindlela ekumele yenze ngayo. Amanyathelo ama-5 okwakhela ii-arhente ze-AI kuqalaUmthetho awusebenzi kwinethiwekhi yoluntu kunye neNetflix :)Bendifuna ukucacisa ii-APIs, izakhono, kunye nee-MCPs kuqala kuba iinkampani ezininzi zicinga ukuba "iarhente-yokuqala" ithetha ukwakha i-MCP.iseva kwaye uyibiza ngokuba yimini.Kodwa i-MCP kufuneka ibe linyathelo lokugqibela.Nantsi amanyathelo ama-5 abalulekileyo, ngokokubaluleka: Musa ukunyanzela abantu ukuba basebenzise iwebhusayithi yakho okanye i-app. Olu lutshintsho lwengqondo ekufuneka luze kuqala. Sifudukela kwihlabathi apho abantu banxibelelana nemveliso yakho ikakhulu nge-ChatGPT, uClaude, okanye enye iarhente ye-AI. Ukuba imveliso yakho isebenza kuphela xa umntu etyelela iwebhusayithi yakho okanye i-app, sele usemva.Imveliso nganye idinga i-API ehambelanayo. Khetha nayiphi na into ebalulekileyo kwimveliso yakho. Ukuba i-arhente ayikwazi ukwenza isenzo esifanayo nge-API, ngoko awuyiyo i-ejenti yokuqala. Ndikhe ndamangaliswa kukuba zingaphi iimveliso ezine-UI ezintle ezihleli phezu kwee-API ezingaphelelanga okanye ezingabhalwanga. Funda ngokugqithisileyo
Kutheni kufuneka Wakhe iMveliso yakho yeearhente ze-AI kuqala
By Creator Economy
·
·
4 min read
·
788 views
Read in:
aa
ace
af
ak
alz
am
ar
as
awa
ay
az
ba
ban
be
bew
+191 more
bg
bho
bik
bm
bn
brx
bs
bug
ca
ceb
cgg
ckb
co
crh
cs
cv
cy
da
de
din
doi
dv
dyu
dz
ee
el
en
eo
es
et
eu
fa
ff
fi
fj
fo
fr
fur
fy
ga
gd
gl
gom
gn
gu
ha
haw
he
hi
hil
hne
hmn
hr
hrx
ht
hu
hy
id
ig
ilo
is
it
ja
jam
jv
ka
kab
kbp
kg
kha
kk
kl
km
kn
ko
kri
ku
ktu
ky
la
lb
lg
li
lij
ln
lo
lmo
lt
ltg
lua
luo
lus
lv
mai
mak
mg
mi
min
mk
ml
mn
mni-mtei
mos
mr
ms
mt
my
nd
ne
nl
nn
no
nr
nso
nus
ny
oc
om
or
pa
pag
pam
pap
pl
ps
pt
pt-br
qu
rn
ro
ru
rw
sa
sah
sat
sc
scn
sg
si
sk
sl
sm
sn
so
sq
sr
ss
st
su
sus
sv
sw
szl
ta
tcy
te
tg
th
ti
tiv
tk
tl
tn
to
tpi
tr
trp
ts
tt
tum
ty
udm
ug
uk
ur
uz
ve
vec
vi
war
wo
xh
yi
yo
yua
yue
zap
zh
zh-hk
zh-tw
zu
