This repository was archived by the owner on Jun 11, 2026. It is now read-only.
-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathcourse.html
More file actions
3288 lines (2887 loc) · 158 KB
/
Copy pathcourse.html
File metadata and controls
3288 lines (2887 loc) · 158 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>GH-600 Exam Prep — Course</title>
<link rel="stylesheet" href="styles.css">
<style>
/* ── Course Layout ── */
.course-layout {
display: grid;
grid-template-columns: 300px 1fr;
gap: 0;
min-height: calc(100vh - 56px);
}
@media (max-width: 1000px) {
.course-layout {
grid-template-columns: 1fr;
}
.course-sidebar {
position: relative !important;
top: 0 !important;
max-height: none !important;
border-right: none !important;
border-bottom: 1px solid var(--border);
}
}
/* ── Sidebar ── */
.course-sidebar {
position: sticky;
top: 56px;
max-height: calc(100vh - 56px);
overflow-y: auto;
background: var(--bg-secondary);
border-right: 1px solid var(--border);
padding: 1.25rem 0;
}
.sidebar-progress {
padding: 0 1.25rem 1.25rem;
border-bottom: 1px solid var(--border);
margin-bottom: 1rem;
}
.sidebar-progress-label {
display: flex;
justify-content: space-between;
align-items: center;
margin-bottom: 0.5rem;
}
.sidebar-progress-label span {
font-size: 0.8rem;
color: var(--text-secondary);
}
.sidebar-progress-label strong {
font-family: var(--font-mono);
font-size: 0.85rem;
color: var(--green);
}
.sidebar-domain {
margin-bottom: 0.25rem;
}
.sidebar-domain-header {
display: flex;
align-items: center;
gap: 0.5rem;
padding: 0.6rem 1.25rem;
cursor: pointer;
font-size: 0.82rem;
font-weight: 600;
color: var(--text-secondary);
transition: all var(--transition);
border: none;
background: none;
width: 100%;
text-align: left;
font-family: var(--font-sans);
}
.sidebar-domain-header:hover {
color: var(--text-primary);
background: var(--bg-card);
}
.sidebar-domain-header.active {
color: var(--green);
background: var(--green-glow);
}
.sidebar-domain-header .domain-indicator {
width: 8px;
height: 8px;
border-radius: 50%;
flex-shrink: 0;
}
.sidebar-domain-header .chevron {
margin-left: auto;
font-size: 0.65rem;
transition: transform 0.2s ease;
flex-shrink: 0;
}
.sidebar-domain-header.expanded .chevron {
transform: rotate(90deg);
}
.sidebar-domain-header .domain-check {
width: 16px;
height: 16px;
border-radius: 50%;
border: 2px solid var(--border);
flex-shrink: 0;
display: flex;
align-items: center;
justify-content: center;
font-size: 0.55rem;
font-weight: 700;
}
.sidebar-domain-header .domain-check.done {
border-color: var(--green);
background: var(--green);
color: var(--bg-primary);
}
.sidebar-subsections {
display: none;
padding: 0.25rem 0 0.5rem 0;
}
.sidebar-subsections.open {
display: block;
}
.sidebar-sub-item {
display: flex;
align-items: center;
gap: 0.5rem;
padding: 0.4rem 1.25rem 0.4rem 2.75rem;
font-size: 0.78rem;
color: var(--text-muted);
cursor: pointer;
transition: all var(--transition);
border: none;
background: none;
width: 100%;
text-align: left;
font-family: var(--font-sans);
}
.sidebar-sub-item:hover {
color: var(--text-primary);
background: var(--bg-card);
}
.sidebar-sub-item.active {
color: var(--green);
}
.sidebar-sub-item .sub-check {
width: 12px;
height: 12px;
border-radius: 50%;
border: 1.5px solid var(--border);
flex-shrink: 0;
}
.sidebar-sub-item.completed .sub-check {
border-color: var(--green);
background: var(--green);
}
/* ── Main Content ── */
.course-main {
padding: 2rem 2.5rem 4rem;
max-width: 900px;
}
@media (max-width: 1000px) {
.course-main { padding: 1.5rem; }
}
.course-header {
margin-bottom: 2.5rem;
padding-bottom: 1.5rem;
border-bottom: 1px solid var(--border);
}
.course-header h1 {
font-size: 1.8rem;
margin-bottom: 0.5rem;
}
.course-header p {
color: var(--text-secondary);
font-size: 0.95rem;
}
/* ── Domain Sections ── */
.domain-section {
margin-bottom: 3rem;
scroll-margin-top: 72px;
}
.domain-title {
display: flex;
align-items: center;
gap: 0.75rem;
margin-bottom: 0.5rem;
}
.domain-title h2 {
font-size: 1.4rem;
}
.domain-title .domain-badge {
font-family: var(--font-mono);
font-size: 0.7rem;
padding: 0.2rem 0.6rem;
border-radius: 4px;
font-weight: 600;
flex-shrink: 0;
}
.domain-meta {
font-size: 0.85rem;
color: var(--text-muted);
margin-bottom: 2rem;
}
/* ── Subsection ── */
.subsection {
margin-bottom: 2.5rem;
scroll-margin-top: 72px;
}
.subsection > h3 {
font-size: 1.15rem;
color: var(--green);
margin-bottom: 1rem;
padding-bottom: 0.5rem;
border-bottom: 1px solid var(--border);
}
.subsection > h4 {
font-size: 1rem;
color: var(--text-primary);
margin: 1.5rem 0 0.75rem;
}
.subsection p {
margin-bottom: 0.75rem;
font-size: 0.92rem;
color: var(--text-secondary);
line-height: 1.7;
}
.subsection ul, .subsection ol {
margin: 0.5rem 0 1rem 1.25rem;
font-size: 0.92rem;
color: var(--text-secondary);
line-height: 1.7;
}
.subsection li {
margin-bottom: 0.35rem;
}
.subsection li strong {
color: var(--text-primary);
}
/* ── Callout Boxes ── */
.exam-tip {
border-left: 3px solid var(--green);
background: var(--green-glow);
padding: 1rem 1.25rem;
border-radius: 0 var(--radius) var(--radius) 0;
margin: 1.25rem 0;
}
.exam-tip-title {
font-weight: 700;
font-size: 0.8rem;
text-transform: uppercase;
letter-spacing: 0.5px;
color: var(--green);
margin-bottom: 0.35rem;
}
.exam-tip p, .exam-tip li {
font-size: 0.88rem;
color: var(--text-secondary);
}
.warning-box {
border-left: 3px solid var(--yellow);
background: rgba(255, 204, 0, 0.08);
padding: 1rem 1.25rem;
border-radius: 0 var(--radius) var(--radius) 0;
margin: 1.25rem 0;
}
.warning-box-title {
font-weight: 700;
font-size: 0.8rem;
text-transform: uppercase;
letter-spacing: 0.5px;
color: var(--yellow);
margin-bottom: 0.35rem;
}
.danger-box {
border-left: 3px solid var(--red);
background: rgba(255, 68, 68, 0.08);
padding: 1rem 1.25rem;
border-radius: 0 var(--radius) var(--radius) 0;
margin: 1.25rem 0;
}
.danger-box-title {
font-weight: 700;
font-size: 0.8rem;
text-transform: uppercase;
letter-spacing: 0.5px;
color: var(--red);
margin-bottom: 0.35rem;
}
.concept-box {
border-left: 3px solid var(--blue);
background: rgba(68, 136, 255, 0.08);
padding: 1rem 1.25rem;
border-radius: 0 var(--radius) var(--radius) 0;
margin: 1.25rem 0;
}
.concept-box-title {
font-weight: 700;
font-size: 0.8rem;
text-transform: uppercase;
letter-spacing: 0.5px;
color: var(--blue);
margin-bottom: 0.35rem;
}
.concept-box p {
font-size: 0.88rem;
color: var(--text-secondary);
}
/* ── Code blocks in course ── */
.subsection pre {
margin: 1rem 0 1.25rem;
}
/* ── Comparison Tables ── */
.subsection table {
margin: 1rem 0 1.5rem;
}
.subsection th {
background: var(--bg-secondary);
}
/* ── Mark Complete Button ── */
.mark-complete-wrapper {
margin-top: 2rem;
padding-top: 1.5rem;
border-top: 1px solid var(--border);
display: flex;
align-items: center;
gap: 1rem;
}
.mark-complete-wrapper .status-text {
font-size: 0.85rem;
color: var(--text-muted);
}
.mark-complete-wrapper .status-text.done {
color: var(--green);
font-weight: 600;
}
.btn-complete {
display: inline-flex;
align-items: center;
gap: 0.5rem;
padding: 0.6rem 1.2rem;
border-radius: var(--radius);
font-size: 0.85rem;
font-weight: 600;
cursor: pointer;
border: 1px solid var(--green-dim);
background: var(--green-glow);
color: var(--green);
font-family: var(--font-sans);
transition: all var(--transition);
}
.btn-complete:hover {
background: var(--green);
color: var(--bg-primary);
box-shadow: 0 0 20px var(--green-glow);
}
.btn-complete.completed {
background: var(--green);
color: var(--bg-primary);
cursor: default;
}
/* ── Diagram boxes ── */
.flow-diagram {
display: flex;
align-items: center;
gap: 0;
flex-wrap: wrap;
margin: 1rem 0 1.5rem;
justify-content: center;
}
.flow-step {
background: var(--bg-card);
border: 1px solid var(--border);
border-radius: var(--radius);
padding: 0.6rem 1rem;
font-size: 0.82rem;
font-weight: 600;
color: var(--text-primary);
text-align: center;
min-width: 100px;
}
.flow-arrow {
color: var(--green);
font-family: var(--font-mono);
font-size: 1.2rem;
padding: 0 0.4rem;
flex-shrink: 0;
}
.flow-step.highlight {
border-color: var(--green);
color: var(--green);
background: var(--green-glow);
}
/* ── Grid layout for comparisons ── */
.compare-grid {
display: grid;
grid-template-columns: 1fr 1fr;
gap: 1rem;
margin: 1rem 0 1.5rem;
}
@media (max-width: 700px) {
.compare-grid { grid-template-columns: 1fr; }
}
.compare-card {
background: var(--bg-card);
border: 1px solid var(--border);
border-radius: var(--radius);
padding: 1rem 1.25rem;
}
.compare-card h5 {
font-size: 0.9rem;
margin-bottom: 0.5rem;
color: var(--text-primary);
}
.compare-card p, .compare-card li {
font-size: 0.85rem;
color: var(--text-secondary);
}
.compare-card ul {
margin: 0.25rem 0 0 1rem;
}
/* ── Footer ── */
.course-footer {
text-align: center;
padding: 2rem;
color: var(--text-muted);
font-size: 0.75rem;
border-top: 1px solid var(--border);
}
</style>
</head>
<body class="scanline">
<nav class="top-nav">
<a href="index.html" class="logo">GH-600<span> // EXAM PREP</span></a>
<div class="nav-links">
<a href="index.html">Hub</a>
<a href="course.html" class="active">Course</a>
<a href="game.html">Game</a>
<a href="exam.html">Exams</a>
<button class="bionic-toggle" id="bionicToggle" title="Toggle Bionic Reading">BIO</button>
</div>
</nav>
<div class="course-layout">
<!-- ════════════════════════════════════════ -->
<!-- SIDEBAR -->
<!-- ════════════════════════════════════════ -->
<aside class="course-sidebar" id="courseSidebar">
<div class="sidebar-progress">
<div class="sidebar-progress-label">
<span>Course Progress</span>
<strong id="progressPercent">0%</strong>
</div>
<div class="progress-bar">
<div class="fill" id="progressBar" style="width: 0%"></div>
</div>
</div>
<!-- Domain 1 -->
<div class="sidebar-domain">
<button class="sidebar-domain-header" data-domain="domain1" aria-expanded="false">
<span class="domain-indicator" style="background: var(--green);"></span>
<span class="domain-check" id="check-domain1"></span>
<span>1. Agent Architecture & SDLC</span>
<span class="chevron">▶</span>
</button>
<div class="sidebar-subsections" id="subs-domain1">
<button class="sidebar-sub-item" data-section="d1s1">
<span class="sub-check" id="subcheck-d1s1"></span>
1.1 Agents in the SDLC
</button>
<button class="sidebar-sub-item" data-section="d1s2">
<span class="sub-check" id="subcheck-d1s2"></span>
1.2 Planning, Reasoning, Action
</button>
<button class="sidebar-sub-item" data-section="d1s3">
<span class="sub-check" id="subcheck-d1s3"></span>
1.3 Observability & Control
</button>
</div>
</div>
<!-- Domain 2 -->
<div class="sidebar-domain">
<button class="sidebar-domain-header" data-domain="domain2" aria-expanded="false">
<span class="domain-indicator" style="background: var(--blue);"></span>
<span class="domain-check" id="check-domain2"></span>
<span>2. Tool Use & Environment</span>
<span class="chevron">▶</span>
</button>
<div class="sidebar-subsections" id="subs-domain2">
<button class="sidebar-sub-item" data-section="d2s1">
<span class="sub-check" id="subcheck-d2s1"></span>
2.1 Select & Configure Tools
</button>
<button class="sidebar-sub-item" data-section="d2s2">
<span class="sub-check" id="subcheck-d2s2"></span>
2.2 Configure MCP Servers
</button>
<button class="sidebar-sub-item" data-section="d2s3">
<span class="sub-check" id="subcheck-d2s3"></span>
2.3 Agent Dev Environments
</button>
<button class="sidebar-sub-item" data-section="d2s4">
<span class="sub-check" id="subcheck-d2s4"></span>
2.4 Safe Execution Paths
</button>
<button class="sidebar-sub-item" data-section="d2s5">
<span class="sub-check" id="subcheck-d2s5"></span>
2.5 Copilot Agent Mode
</button>
<button class="sidebar-sub-item" data-section="d2s6">
<span class="sub-check" id="subcheck-d2s6"></span>
2.6 Copilot Extensions
</button>
<button class="sidebar-sub-item" data-section="d2s7">
<span class="sub-check" id="subcheck-d2s7"></span>
2.7 Agent Skills
</button>
<button class="sidebar-sub-item" data-section="d2s8">
<span class="sub-check" id="subcheck-d2s8"></span>
2.8 Copilot SDK
</button>
</div>
</div>
<!-- Domain 3 -->
<div class="sidebar-domain">
<button class="sidebar-domain-header" data-domain="domain3" aria-expanded="false">
<span class="domain-indicator" style="background: var(--purple);"></span>
<span class="domain-check" id="check-domain3"></span>
<span>3. Memory, State & Execution</span>
<span class="chevron">▶</span>
</button>
<div class="sidebar-subsections" id="subs-domain3">
<button class="sidebar-sub-item" data-section="d3s1">
<span class="sub-check" id="subcheck-d3s1"></span>
3.1 Agent Memory Strategies
</button>
<button class="sidebar-sub-item" data-section="d3s2">
<span class="sub-check" id="subcheck-d3s2"></span>
3.2 Persist State & Context Drift
</button>
<button class="sidebar-sub-item" data-section="d3s3">
<span class="sub-check" id="subcheck-d3s3"></span>
3.3 Cross-Tool Continuity
</button>
</div>
</div>
<!-- Domain 4 -->
<div class="sidebar-domain">
<button class="sidebar-domain-header" data-domain="domain4" aria-expanded="false">
<span class="domain-indicator" style="background: var(--yellow);"></span>
<span class="domain-check" id="check-domain4"></span>
<span>4. Evaluation & Tuning</span>
<span class="chevron">▶</span>
</button>
<div class="sidebar-subsections" id="subs-domain4">
<button class="sidebar-sub-item" data-section="d4s1">
<span class="sub-check" id="subcheck-d4s1"></span>
4.1 Success Criteria & Signals
</button>
<button class="sidebar-sub-item" data-section="d4s2">
<span class="sub-check" id="subcheck-d4s2"></span>
4.2 Failure & Root Cause Analysis
</button>
<button class="sidebar-sub-item" data-section="d4s3">
<span class="sub-check" id="subcheck-d4s3"></span>
4.3 Tune Agent Behavior
</button>
</div>
</div>
<!-- Domain 5 -->
<div class="sidebar-domain">
<button class="sidebar-domain-header" data-domain="domain5" aria-expanded="false">
<span class="domain-indicator" style="background: var(--orange);"></span>
<span class="domain-check" id="check-domain5"></span>
<span>5. Multi-Agent Coordination</span>
<span class="chevron">▶</span>
</button>
<div class="sidebar-subsections" id="subs-domain5">
<button class="sidebar-sub-item" data-section="d5s1">
<span class="sub-check" id="subcheck-d5s1"></span>
5.1 Multi-Agent Workflows
</button>
<button class="sidebar-sub-item" data-section="d5s2">
<span class="sub-check" id="subcheck-d5s2"></span>
5.2 Multi-Agent Observability
</button>
<button class="sidebar-sub-item" data-section="d5s3">
<span class="sub-check" id="subcheck-d5s3"></span>
5.3 Multi-Agent Failures
</button>
<button class="sidebar-sub-item" data-section="d5s4">
<span class="sub-check" id="subcheck-d5s4"></span>
5.4 Agent Lifecycle Management
</button>
</div>
</div>
<!-- Domain 6 -->
<div class="sidebar-domain">
<button class="sidebar-domain-header" data-domain="domain6" aria-expanded="false">
<span class="domain-indicator" style="background: var(--red);"></span>
<span class="domain-check" id="check-domain6"></span>
<span>6. Guardrails & Accountability</span>
<span class="chevron">▶</span>
</button>
<div class="sidebar-subsections" id="subs-domain6">
<button class="sidebar-sub-item" data-section="d6s1">
<span class="sub-check" id="subcheck-d6s1"></span>
6.1 Define Autonomy Levels
</button>
<button class="sidebar-sub-item" data-section="d6s2">
<span class="sub-check" id="subcheck-d6s2"></span>
6.2 Guardrails & HITL
</button>
</div>
</div>
</aside>
<!-- ════════════════════════════════════════ -->
<!-- MAIN CONTENT -->
<!-- ════════════════════════════════════════ -->
<main class="course-main" id="courseMain">
<div class="course-header">
<h1>GH-600 <span class="text-green">Study Guide</span></h1>
<p>Complete walkthrough of all 6 exam domains. Work through each section, then mark it complete to track your progress.</p>
</div>
<!-- ================================================================== -->
<!-- DOMAIN 1: Prepare Agent Architecture & SDLC Processes (15-20%) -->
<!-- ================================================================== -->
<section class="domain-section" id="domain1">
<div class="domain-title">
<span class="domain-badge badge-green">15-20%</span>
<h2>Domain 1: Prepare Agent Architecture & SDLC Processes</h2>
</div>
<p class="domain-meta">3 subtopics — Understand what makes software agents "agentic," how they integrate into the software development lifecycle, and how to configure observability and control over their autonomous behavior.</p>
<!-- ── 1.1 ── -->
<div class="subsection" id="d1s1">
<h3>1.1 Integrate Agents into the SDLC</h3>
<h4>What Makes an AI System "Agentic" vs. "Assistant"</h4>
<p>The exam draws a clear line between an <strong>assistant</strong> and an <strong>agent</strong>. An assistant responds to a single prompt — you ask it to write a function, it writes a function, and it's done. An agent is fundamentally different: it maintains <strong>goals across multiple steps</strong>, decides which actions to take, uses tools to interact with external systems, creates <strong>durable artifacts</strong> (branches, commits, pull requests, files), and iterates on feedback from those actions.</p>
<div class="compare-grid">
<div class="compare-card">
<h5>Assistant (Non-Agentic)</h5>
<ul>
<li>Single prompt → single response</li>
<li>Suggests code, does not execute it</li>
<li>No tool use beyond generation</li>
<li>No persistent state between turns</li>
<li>User manually applies suggestions</li>
</ul>
</div>
<div class="compare-card">
<h5>Agent (Agentic)</h5>
<ul>
<li>Maintains goals across many steps</li>
<li>Decides <em>which</em> actions to take next</li>
<li>Uses tools: file I/O, GitHub API, shell, MCP</li>
<li>Creates durable artifacts (branches, PRs)</li>
<li>Iterates based on feedback (CI, reviews)</li>
</ul>
</div>
</div>
<div class="exam-tip">
<div class="exam-tip-title">Key Exam Concept</div>
<p>The defining characteristic of an <strong>agentic</strong> system is its ability to <strong>autonomously decide and execute multi-step actions</strong> toward a goal, creating lasting changes in the codebase. If it only suggests — it's an assistant. If it acts — it's an agent.</p>
</div>
<h4>Steps Agents Can Perform in the SDLC</h4>
<p>Agents can participate in nearly every phase of the software development lifecycle. The exam expects you to know which phases are well-suited for agent automation and which require human oversight:</p>
<ul>
<li><strong>Code generation:</strong> Creating new files, writing functions, scaffolding entire features from issue descriptions. The agent reads the issue, plans the implementation, writes code, and pushes it to a feature branch.</li>
<li><strong>Code review:</strong> An agent can review PRs for style consistency, potential bugs, missing tests, or security issues. It leaves inline comments just like a human reviewer.</li>
<li><strong>Testing:</strong> Generating unit tests, integration tests, running test suites, and reporting results. Agents can identify untested code paths and write tests to cover them.</li>
<li><strong>Dependency updates:</strong> Scanning for outdated dependencies, creating PRs with version bumps, running tests to verify compatibility (similar to what Dependabot does, but with agentic reasoning about breaking changes).</li>
<li><strong>Deployment:</strong> Triggering deployment pipelines, monitoring rollout health, performing rollbacks if metrics degrade.</li>
<li><strong>Documentation:</strong> Generating or updating docs based on code changes, keeping READMEs in sync, writing changelog entries.</li>
</ul>
<h4>Common Anti-Patterns</h4>
<p>The exam tests your ability to identify when agent usage goes wrong. Memorize these anti-patterns:</p>
<table>
<thead>
<tr><th>Anti-Pattern</th><th>Description</th><th>Mitigation</th></tr>
</thead>
<tbody>
<tr>
<td><strong>Over-autonomy</strong></td>
<td>Agent acts without any human review gates — merges PRs, deploys to production without checks.</td>
<td>Required approvals, branch protection, status checks before merge.</td>
</tr>
<tr>
<td><strong>Context pollution</strong></td>
<td>Stale or conflicting context accumulates, causing the agent to make decisions based on outdated information.</td>
<td>Context pruning, freshness checks, re-reading current state before acting.</td>
</tr>
<tr>
<td><strong>Goal drift</strong></td>
<td>Agent gradually diverges from the original task — starts fixing "one more thing," refactoring unrelated code.</td>
<td>Scoped task definitions, explicit success criteria, boundary enforcement.</td>
</tr>
<tr>
<td><strong>Missing traceability</strong></td>
<td>No audit trail of what the agent did or why — commits lack context, no PR descriptions, no linked issues.</td>
<td>Require meaningful commit messages, PR descriptions, issue linking.</td>
</tr>
<tr>
<td><strong>Scope creep</strong></td>
<td>Agent exceeds the boundaries of its assigned task — modifies files outside its scope, changes configuration it shouldn't touch.</td>
<td>File/directory scoping, CODEOWNERS, restricted tool permissions.</td>
</tr>
</tbody>
</table>
<div class="warning-box">
<div class="warning-box-title">Anti-Pattern Warning</div>
<p><strong>Over-autonomy</strong> and <strong>missing traceability</strong> are the two most dangerous anti-patterns in production. Every agent action should be traceable to a commit, PR, or workflow run. Every high-impact action should have a human review gate.</p>
</div>
<h4>Defining Inputs, Outputs, and Success Criteria</h4>
<p>A well-configured agent has clearly defined boundaries for each task:</p>
<ul>
<li><strong>Inputs:</strong> The issue description, requirements documents, context files (like <code>copilot-instructions.md</code>), existing code for reference, test fixtures, environment configuration.</li>
<li><strong>Outputs:</strong> Feature branches, pull requests, code changes, test files, documentation updates, SARIF scan results, deployment artifacts.</li>
<li><strong>Success criteria:</strong> All tests pass, no new security vulnerabilities (clean CodeQL/secret scanning), code review approved, performance benchmarks met, coverage thresholds maintained. These should be defined <em>before</em> the agent starts work.</li>
</ul>
<pre><code># Example: Well-defined agent task specification
Task: Implement user authentication endpoint
Inputs:
- Issue #142 (requirements)
- /src/api/ (existing API patterns)
- /tests/api/ (test patterns)
Outputs:
- Branch: feature/auth-endpoint
- PR with description and linked issue
- Source files in /src/api/auth/
- Tests in /tests/api/auth/
Success Criteria:
- All existing tests still pass
- New endpoint has >= 80% test coverage
- CodeQL scan: zero new alerts
- Secret scanning: clean
- PR approved by CODEOWNERS</code></pre>
<div class="mark-complete-wrapper">
<button class="btn-complete" data-section="d1s1">Mark as Complete</button>
<span class="status-text" id="status-d1s1"></span>
</div>
</div>
<!-- ── 1.2 ── -->
<div class="subsection" id="d1s2">
<h3>1.2 Define Boundaries Between Planning, Reasoning, and Action</h3>
<h4>The Plan → Act → Evaluate Lifecycle Loop</h4>
<p>Agentic workflows are <strong>not linear</strong>. They follow a cyclical loop:</p>
<div class="flow-diagram">
<div class="flow-step highlight">Plan</div>
<span class="flow-arrow">→</span>
<div class="flow-step">Act</div>
<span class="flow-arrow">→</span>
<div class="flow-step">Evaluate</div>
<span class="flow-arrow">→</span>
<div class="flow-step highlight">Plan</div>
<span class="flow-arrow">...</span>
</div>
<p>The agent plans what to do, acts (writes code, creates a PR), evaluates (reads CI results, review feedback), and then plans again based on that evaluation. This cycle continues until the task meets its success criteria or the agent escalates to a human.</p>
<h4>Planning Artifacts</h4>
<p>When an agent plans, it should produce <strong>inspectable planning artifacts</strong> that humans can review:</p>
<ul>
<li><strong>Structured plans in PR descriptions:</strong> A checklist of what the agent intends to do, with scope, success criteria, and rollback paths.</li>
<li><strong>Linked issues:</strong> The plan references the issue(s) it's addressing, maintaining traceability.</li>
<li><strong>Checklists with scope/success criteria/rollback paths:</strong> Explicitly stating what's in-scope, what done looks like, and what to do if things go wrong.</li>
</ul>
<pre><code>## Agent Plan — PR #347
### Scope
- Add rate limiting middleware to /api/v2/* endpoints
- Update existing tests to cover rate limit responses
### Success Criteria
- [ ] Rate limit returns 429 with Retry-After header
- [ ] Existing API tests pass
- [ ] New rate limit tests have > 90% branch coverage
- [ ] CodeQL scan clean
### Rollback Path
- Revert this PR (no database migrations involved)
- Rate limiting config is env-var driven, can be disabled via RATE_LIMIT_ENABLED=false</code></pre>
<h4>Acting: What Agents Do</h4>
<p>The "Act" phase involves the agent making <strong>durable changes</strong>:</p>
<ul>
<li>Creating branches (<code>git checkout -b feature/rate-limit</code>)</li>
<li>Changing files (writing, editing, deleting source code)</li>
<li>Pushing commits (with meaningful commit messages)</li>
<li>Opening or updating pull requests</li>
<li>Responding to review feedback (addressing comments, pushing fixes)</li>
</ul>
<h4>Evaluating: How Agents Learn From Results</h4>
<p>After acting, the agent evaluates by reading signals from the development environment:</p>
<ul>
<li><strong>Workflow runs:</strong> Did the CI pipeline pass? Which tests failed?</li>
<li><strong>Status checks:</strong> Are all required checks green?</li>
<li><strong>Code review feedback:</strong> What did reviewers say? Are there requested changes?</li>
<li><strong>Security signals:</strong> Code scanning results (CodeQL), secret scanning alerts, dependency vulnerability alerts (Dependabot), SARIF uploads from third-party tools.</li>
</ul>
<div class="concept-box">
<div class="concept-box-title">Key Exam Concept</div>
<p><strong>SARIF (Static Analysis Results Interchange Format)</strong> is the standard format for reporting static analysis results to GitHub. When an agent runs a scanner (CodeQL or any third-party tool), the results are uploaded as SARIF files to GitHub's code scanning API. The agent then reads these results as evaluation signals.</p>
</div>
<h4>Option A: Plan-First PR vs. Option B: Plan + Execution in Same PR</h4>
<div class="compare-grid">
<div class="compare-card">
<h5>Option A: Plan-First PR</h5>
<p>The agent opens a PR with <strong>only a plan</strong> — no code changes. A human reviews and approves the plan. Only after approval does the agent implement.</p>
<ul>
<li>Best for <strong>high-risk work</strong> (schema changes, security-sensitive code, infrastructure)</li>
<li>Adds latency — waiting for plan approval</li>
<li>Maximum safety and alignment</li>
</ul>
</div>
<div class="compare-card">
<h5>Option B: Plan + Execution in Same PR</h5>
<p>The agent includes both the plan (in the PR description) and the implementation (in the code changes) in a single PR.</p>
<ul>
<li>Best for <strong>low-risk, fast iteration</strong> (bug fixes, dependency bumps, documentation)</li>
<li>Faster turnaround</li>
<li>Reviewer evaluates plan and code together</li>
</ul>
</div>
</div>
<h4>Enforcing Planning Boundaries</h4>
<p>How do you prevent an agent from skipping the planning phase and jumping straight to code?</p>
<ul>
<li><strong>Read-only planning agents:</strong> A dedicated agent with only read access to the repository. It can analyze the issue and produce a plan, but it physically cannot write code. A separate agent with write access implements the plan.</li>
<li><strong>Explicit handoff to implementation agents:</strong> The planning agent produces a structured plan and creates an issue or updates a PR description. A different agent (or the same agent in a different mode) picks it up for implementation.</li>
<li><strong>Tool gating in orchestrators:</strong> The orchestration layer only grants write tools (file editing, git push) after a planning phase is complete and reviewed.</li>
<li><strong>"Plan mode" workflows:</strong> The agent is configured via <code>copilot-instructions.md</code> to always output a plan first and wait for approval before proceeding.</li>
</ul>
<h4>Configuring Agent Planning Behavior</h4>
<p>You configure how Copilot and other agents handle planning through instruction files:</p>
<pre><code># .github/copilot-instructions.md
## Planning Requirements
- Before writing any code, output a structured plan as a PR comment
- Include: scope, files to modify, success criteria, rollback path
- Wait for a human to approve the plan by commenting "approved"
- If the plan is rejected, revise based on feedback and resubmit
- Never skip the planning phase for changes touching /src/core/ or /infrastructure/</code></pre>
<div class="exam-tip">
<div class="exam-tip-title">Key Exam Concept</div>
<p>The file <code>.github/copilot-instructions.md</code> is the <strong>repo-level configuration</strong> for GitHub Copilot agent behavior. It's the primary mechanism for controlling how the agent plans, acts, and self-constrains. Expect questions about what belongs in this file.</p>
</div>
<h4>Validating Agent Plans Before Execution</h4>
<p>Validation can happen through multiple mechanisms:</p>
<ul>
<li><strong>Human review:</strong> A required approval on the plan before the agent proceeds.</li>
<li><strong>Automated checks:</strong> A workflow that validates the plan structure (does it include rollback paths? does it specify success criteria?).</li>
<li><strong>Scope validation:</strong> Ensuring the plan doesn't touch files outside its designated scope.</li>
<li><strong>Risk assessment:</strong> Automated classification of the plan's risk level, routing high-risk plans for additional review.</li>
</ul>
<div class="mark-complete-wrapper">
<button class="btn-complete" data-section="d1s2">Mark as Complete</button>
<span class="status-text" id="status-d1s2"></span>
</div>
</div>
<!-- ── 1.3 ── -->
<div class="subsection" id="d1s3">
<h3>1.3 Configure Observability and Control for Autonomous Agents</h3>
<h4>Autonomy Levels</h4>
<p>The exam defines a spectrum of autonomy. You need to match the right autonomy level to the situation. These map directly to GitHub Copilot's interaction modes:</p>
<table>
<thead>
<tr><th>Level</th><th>Copilot Mode</th><th>Description</th><th>When to Use</th></tr>
</thead>
<tbody>
<tr>
<td><strong>Fully autonomous</strong></td>
<td>Agent + auto-merge</td>
<td>Agent acts, merges, deploys without any human interaction.</td>
<td>Only for extremely low-risk, well-tested, reversible operations (e.g., auto-fixing lint errors).</td>
</tr>
<tr>
<td><strong>Supervised autonomous</strong></td>
<td>Agent mode (standard)</td>
<td>Agent acts autonomously but a human reviews output before merge/deploy.</td>
<td>Standard development tasks — feature PRs, bug fixes, dependency updates. <strong>Most common level.</strong></td>
</tr>
<tr>
<td><strong>Human-in-the-loop</strong></td>
<td>Agent mode with approval gates</td>
<td>Agent proposes, human approves at key decision points during execution.</td>
<td>Higher-risk tasks — schema changes, infrastructure modifications, security-related code.</td>
</tr>
<tr>
<td><strong>Human-led</strong></td>
<td>Ask mode</td>
<td>Human drives the process; agent assists with specific sub-tasks on request.</td>
<td>Novel, unprecedented, or compliance-critical work where the agent lacks sufficient context.</td>
</tr>