Buckets:

HuggingFaceDocBuilder's picture
download
raw
6.43 kB
import{s as O,n as R,o as B}from"../chunks/scheduler.b8c17244.js";import{S as I,i as J,e as Q,s,c as l,h as K,a as L,d as n,b as i,f as D,g as o,j as V,k as G,l as W,m as a,n as u,t as p,o as f,p as m}from"../chunks/index.d374165a.js";import{C as X,H as Y,E as Z}from"../chunks/MermaidChart.svelte_svelte_type_style_lang.200f62db.js";import{Q as q}from"../chunks/Question.b45f7d7c.js";function ee(j){let r,S,v,P,g,_,c,C,h,H="Test your understanding of subagent patterns, architectures, and when to use them.",z,$,F,x,N,b,T,d,A,w,E,k,U,y,M;return g=new X({props:{containerStyle:"float: right; margin-left: 10px; display: inline-flex; position: relative; z-index: 10;"}}),c=new Y({props:{title:"Quiz 1: Subagent Concepts",local:"quiz-1-subagent-concepts",headingTag:"h1"}}),$=new q({props:{choices:[{text:"A subagent is a different agent platform (like switching from Claude Code to Codex)",explain:"No. A subagent is a child process spawned by a parent agent within the same platform."},{text:"A subagent is a skill that runs in the background while the main agent works on something else",explain:"Not quite. Subagents are full agents, not just skills. Skills are reusable workflows."},{text:"A subagent is a child agent process spawned by a parent agent to handle a subtask in isolation, then report results back",explain:"Correct! Subagents are independent processes that solve specific subtasks and communicate results to the parent.",correct:!0},{text:"A subagent is a way to run the same agent twice for redundancy",explain:"No. Subagents are for dividing work, not duplicating it."}]}}),x=new q({props:{choices:[{text:"Fan-out/fan-in spawns multiple subagents, each handling a sequential stage; each stage waits for the previous to finish",explain:"That's a pipeline pattern, not fan-out/fan-in. Fan-out/fan-in runs all subagents in parallel."},{text:"Fan-out/fan-in spawns multiple independent subagents in parallel, waits for all to finish, then combines results",explain:"Correct! All work happens simultaneously. Perfect for independent tasks.",correct:!0},{text:"Fan-out/fan-in requires a supervisor agent to coordinate the subagents",explain:"Not necessarily. The parent agent can manage it with wait_all() or similar."},{text:"Fan-out/fan-in is the only pattern you should ever use",explain:"No. Different patterns suit different task structures. Pipeline and supervisor are also valuable."}]}}),b=new q({props:{choices:[{text:"Pipeline pattern runs all subagents in parallel to maximize speed",explain:"No, that's fan-out/fan-in. Pipeline runs stages sequentially: output of stage 1 → input of stage 2."},{text:"Pipeline pattern chains subagents sequentially, where each stage's output becomes the next stage's input",explain:"Correct! Pipelines are for sequential workflows like design → code → test.",correct:!0},{text:"Pipeline pattern requires 10+ subagents to be effective",explain:"No. Pipelines work with any number of stages, often 3-5 is ideal."},{text:"Pipeline and fan-out/fan-in are the same thing",explain:"False. Fan-out/fan-in is parallel. Pipeline is sequential."}]}}),d=new q({props:{choices:[{text:"Use subagents when you have 10+ files to read or 3+ independent pieces of work",explain:"Correct! The blog post identifies these as the strong signals for subagent use.",correct:!0},{text:"Always use subagents if your task takes more than 1 minute",explain:"False. Time doesn't matter. Subagent overhead kills small tasks."},{text:"Use subagents only for parallel work; never for sequential tasks",explain:"False. Pipeline pattern uses subagents for sequential work."},{text:"Use subagents for every task; they always make things faster",explain:"No. Subagent spawn overhead makes them slower for small tasks."}]}}),w=new q({props:{choices:[{text:"Supervisor pattern spawns all subagents in parallel to maximize speed",explain:"That's fan-out/fan-in, not supervisor. Supervisor uses specialized agents with different tools."},{text:"Supervisor pattern has a parent agent directing multiple specialist subagents, each with different tools and expertise",explain:"Correct! Supervisor is for when you need multiple specialized agents working on the same problem.",correct:!0},{text:"Supervisor pattern requires more than 10 subagents to be effective",explain:"No. Usually 2-5 specialist subagents is ideal with supervisor pattern."},{text:"Supervisor pattern means all subagents have identical tools",explain:"False. Supervisor specifically gives each subagent specialized tools for its domain."}]}}),k=new Z({props:{source:"https://github.com/huggingface/context-course/blob/main/units/en/unit4/quiz1.mdx"}}),{c(){r=Q("meta"),S=s(),v=Q("p"),P=s(),l(g.$$.fragment),_=s(),l(c.$$.fragment),C=s(),h=Q("p"),h.textContent=H,z=s(),l($.$$.fragment),F=s(),l(x.$$.fragment),N=s(),l(b.$$.fragment),T=s(),l(d.$$.fragment),A=s(),l(w.$$.fragment),E=s(),l(k.$$.fragment),U=s(),y=Q("p"),this.h()},l(e){const t=K("svelte-u9bgzb",document.head);r=L(t,"META",{name:!0,content:!0}),t.forEach(n),S=i(e),v=L(e,"P",{}),D(v).forEach(n),P=i(e),o(g.$$.fragment,e),_=i(e),o(c.$$.fragment,e),C=i(e),h=L(e,"P",{"data-svelte-h":!0}),V(h)!=="svelte-21lkip"&&(h.textContent=H),z=i(e),o($.$$.fragment,e),F=i(e),o(x.$$.fragment,e),N=i(e),o(b.$$.fragment,e),T=i(e),o(d.$$.fragment,e),A=i(e),o(w.$$.fragment,e),E=i(e),o(k.$$.fragment,e),U=i(e),y=L(e,"P",{}),D(y).forEach(n),this.h()},h(){G(r,"name","hf:doc:metadata"),G(r,"content",te)},m(e,t){W(document.head,r),a(e,S,t),a(e,v,t),a(e,P,t),u(g,e,t),a(e,_,t),u(c,e,t),a(e,C,t),a(e,h,t),a(e,z,t),u($,e,t),a(e,F,t),u(x,e,t),a(e,N,t),u(b,e,t),a(e,T,t),u(d,e,t),a(e,A,t),u(w,e,t),a(e,E,t),u(k,e,t),a(e,U,t),a(e,y,t),M=!0},p:R,i(e){M||(p(g.$$.fragment,e),p(c.$$.fragment,e),p($.$$.fragment,e),p(x.$$.fragment,e),p(b.$$.fragment,e),p(d.$$.fragment,e),p(w.$$.fragment,e),p(k.$$.fragment,e),M=!0)},o(e){f(g.$$.fragment,e),f(c.$$.fragment,e),f($.$$.fragment,e),f(x.$$.fragment,e),f(b.$$.fragment,e),f(d.$$.fragment,e),f(w.$$.fragment,e),f(k.$$.fragment,e),M=!1},d(e){e&&(n(S),n(v),n(P),n(_),n(C),n(h),n(z),n(F),n(N),n(T),n(A),n(E),n(U),n(y)),n(r),m(g,e),m(c,e),m($,e),m(x,e),m(b,e),m(d,e),m(w,e),m(k,e)}}}const te='{"title":"Quiz 1: Subagent Concepts","local":"quiz-1-subagent-concepts","sections":[],"depth":1}';function ne(j){return B(()=>{new URLSearchParams(window.location.search).get("fw")}),[]}class le extends I{constructor(r){super(),J(this,r,ne,ee,O,{})}}export{le as component};

Xet Storage Details

Size:
6.43 kB
·
Xet hash:
3f9a3e5e2dd0bcb0ca927d5c3d0a0fe0297718c240f32a90d06d1e66d3c1effb

Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.