- Preface i
-
1 What an Agentic Coding Harness Is 2
- 1.1 The Moment Coding Tools Stopped Being Just Assistants 2
- 1.2 Why ``AI Coding Tool'' Is Too Vague 3
- 1.3 The Developer Tooling Spectrum 4
- 1.4 Static Autocomplete: Prediction Without Agency 5
- 1.5 Chat Assistants: Reasoning Without Direct Execution 5
- 1.6 Inline Chat and IDE Assistants: Context-Aware Help 7
- 1.7 Agentic Coding Harnesses: The Model Gets Hands 7
- 1.8 A Precise Definition of an Agentic Coding Harness 8
- 1.9 The Four Capabilities of a True Harness 9
- 1.10 Capability 1: Autonomous File-System Read and Write 10
- 1.11 Capability 2: Shell Command Execution 10
- 1.12 Capability 3: Recursive Self-Correction 11
- 1.13 Capability 4: Tool Registration and Tool Schemas 12
- 1.14 Human-in-the-Loop vs. Human-on-the-Loop 13
- 1.15 Why Harnesses Need More Permissions Than Chat 14
- 1.16 What Harnesses Still Cannot Safely Do 15
- 1.17 Tool Calls as the Boundary Between Text and Action 15
- 1.18 Walkthrough: Adding Input Validation with a Harness 16
- 1.19 Comparing the Same Task Across Tool Categories 18
- 1.20 How to Classify Any New AI Coding Tool 19
- 1.21 Common Misconceptions About Coding Agents 20
- 1.22 Hands-On Lab: First Agentic Session 21
- 1.23 Interview Questions 22
- 1.24 Chapter Summary 24
- 1.25 What Comes Next 25
-
2 The Agent Loop 26
- 2.1 The Loop Behind the Illusion 27
- 2.2 Why Harnesses Are Easier to Debug Once You See the Loop 27
- 2.3 A Tiny Tool-Call Round Trip 29
- 2.4 The Five Phases of the Agent Loop 30
- 2.5 Phase 1: Perception 31
- 2.6 Phase 2: Planning 32
- 2.7 Phase 3: Execution 33
- 2.8 Phase 4: Observation 34
- 2.9 Phase 5: Termination 35
- 2.10 Correction: The Loop Learns From Its Own Tool Output 35
- 2.11 Tool Calls as Structured Intent 36
- 2.12 What the Harness, Not the Model, Actually Executes 37
- 2.13 How Context Is Rebuilt on Every Turn 38
- 2.14 Turn Limits, Token Budgets, and Timeouts 38
- 2.15 Permission Denials as Loop Events 39
- 2.16 Failed Commands and Error Re-Injection 40
- 2.17 Clean Termination vs. Stalled Termination 41
- 2.18 The Loop in Pseudocode 42
- 2.19 Walkthrough: A Docstring Task From Start to Finish 43
- 2.20 Reading Agent Logs 46
- 2.21 Diagnosing Common Loop Failures 47
- 2.22 How Different Harnesses Expose the Loop 49
- 2.23 Hands-On Lab: Trace a Real Agent Loop 49
- 2.24 Interview Questions 51
- 2.25 Chapter Summary 53
- 2.26 What Comes Next 54
-
3 Anatomy of a Harness 55
- 3.1 From Loop Behavior to Harness Architecture 55
- 3.2 The Five Components Every Harness Needs 56
- 3.3 Why Anatomy Matters When Agents Misbehave 58
- 3.4 Component 1: The System Prompt 58
- 3.5 System Prompt Layering: Built-In, Global, Project, Session 59
- 3.6 Global Instruction Files 61
- 3.7 Project-Level Instruction Files: AGENTS.md and CLAUDE.md 62
- 3.8 What Belongs in Project Instructions 63
- 3.9 What Does Not Belong in Project Instructions 64
- 3.10 Component 2: Tool Definitions 65
- 3.11 Tool Schemas, Parameters, and Descriptions 66
- 3.12 Built-In Tools vs. Registered Tools 67
- 3.13 MCP as a Tool-Registration Layer 68
- 3.14 Component 3: The Context and State Manager 69
- 3.15 How a Harness Decides What the Model Sees 70
- 3.16 Conversation History, File Contents, Repo Maps, and Summaries 71
- 3.17 Context Drift and Context Pollution 71
- 3.18 Component 4: The Permission Gate 72
- 3.19 Approval, Denial, Confirmation, and Auditability 73
- 3.20 Component 5: The LLM Client 75
- 3.21 Provider APIs, Local Endpoints, and Model Routing 75
- 3.22 Mapping Configuration Files to Harness Components 77
- 3.23 Troubleshooting by Component 78
- 3.24 Hands-On Lab: Build Your First Harness Anatomy Map 80
- 3.25 Interview Questions 81
- 3.26 Chapter Summary 83
- 3.27 What Comes Next 84
-
4 The Model Context Protocol (MCP) 87
- 4.1 Why Harnesses Need External Tools 87
- 4.2 The Problem MCP Solves 88
- 4.3 MCP in One Sentence 89
- 4.4 Clients, Servers, Tools, and Results 90
- 4.5 How MCP Fits Into the Agent Loop 91
- 4.6 Tool Discovery: How the Harness Learns What Exists 92
- 4.7 Tool Schemas: How Capabilities Become Callable 93
- 4.8 Tool Calls: From Model Intent to Server Request 94
- 4.9 Tool Results: Observations Returned to the Loop 95
- 4.10 Transport Option 1: stdio 96
- 4.11 Transport Option 2: HTTP/SSE 97
- 4.12 Choosing stdio vs. HTTP/SSE 97
- 4.13 Registering a Fictional MCP Server 99
- 4.14 Building a Minimal Python MCP Server 100
- 4.15 Validating Tool Inputs 101
- 4.16 Returning Useful Tool Results 103
- 4.17 Error Handling and Recovery 103
- 4.18 Security: Every Tool Is a Capability 104
- 4.19 Authentication and Authorization for Networked MCP 105
- 4.20 Logging and Auditability 106
- 4.21 Common MCP Failure Modes 107
- 4.22 Hands-On Lab: Build a Tiny todo_mcp_server 109
- 4.23 Interview Questions 110
- 4.24 Chapter Summary 112
- 4.25 What Comes Next 113
-
5 Context & Cache Management 114
- 5.1 When Good Agent Sessions Go Stale 114
- 5.2 What ``Context'' Means in a Coding Harness 115
- 5.3 The Context Window Is a Budget, Not a Filing Cabinet 116
- 5.4 What Enters the Context on Each Turn 117
- 5.5 How Context Grows During an Agent Loop 118
- 5.6 Input Tokens, Output Tokens, and Tool Results 119
- 5.7 Why More Context Is Not Always Better 120
- 5.8 Context Pollution: When Observations Become Noise 121
- 5.9 Prompt Caching: Paying Less for Stable Prefixes 122
- 5.10 What Caching Does Not Solve 123
- 5.11 Compaction: Compressing the Session State 123
- 5.12 What Compaction Can Lose 124
- 5.13 Clearing and Restarting Sessions 125
- 5.14 Repo Maps and Symbol-Level Context 126
- 5.15 Full Files vs. Snippets vs. Summaries 127
- 5.16 MCP Results and Context Pressure 129
- 5.17 Task Scoping as the Primary Cost Control 130
- 5.18 Estimating Token Growth Before You Run 131
- 5.19 Cost Formulas Without Hardcoded Prices 132
- 5.20 Local Models and Smaller Context Windows 133
- 5.21 Walkthrough: Rescuing a Drifting Session 134
- 5.22 Context Hygiene Checklist 135
- 5.23 Hands-On Lab: Measure Context Growth 138
- 5.24 Interview Questions 139
- 5.25 Chapter Summary 140
- 5.26 What Comes Next 142
-
6 Tool Design & Safety 143
- 6.1 Tools Turn Text Into Effects 144
- 6.2 Why Tool Design Is a Safety Problem 145
- 6.3 The Standard Built-In Tool Set 146
- 6.4 File Read Tools 147
- 6.5 Preventing Path Traversal 148
- 6.6 File Write and Edit Tools 149
- 6.7 Diffs, Atomic Writes, and Reviewability 151
- 6.8 Search Tools and Their Failure Modes 151
- 6.9 Shell Execution: The Sharpest Tool 152
- 6.10 Dangerous Command Patterns 153
- 6.11 Test Runner Tools as Safer Shell Alternatives 155
- 6.12 Web Fetch and Network-Aware Tools 156
- 6.13 MCP Tools as Capability Surfaces 157
- 6.14 Permission Gates: Allow, Confirm, Deny 158
- 6.15 Pattern-Based Command Policies 159
- 6.16 Path-Based Permission Rules 160
- 6.17 Session Decisions and Audit Logs 161
- 6.18 Sandboxing Beyond Permission Prompts 162
- 6.19 Read-Only Workspaces and Writable Scratch Areas 163
- 6.20 Prompt Injection Through Repository Content 164
- 6.21 Linguistic Defenses vs. Hard Boundaries 165
- 6.22 Least-Privilege Tool Design 166
- 6.23 Returning Safe and Useful Tool Results 167
- 6.24 Walkthrough: Blocking a Risky Cleanup Command 167
- 6.25 Tool Safety Checklist 169
- 6.26 Hands-On Lab: Test a Permission Policy 170
- 6.27 Interview Questions 171
- 6.28 Chapter Summary 173
- 6.29 What Comes Next 174
-
7 Claude Code Deep Dive 177
- 7.1 Why Claude Code Gets the First Deep Dive 177
- 7.2 Claude Code Through the Harness Anatomy 178
- 7.3 Installation and First Run, Without Overfitting to Today's Syntax 180
- 7.4 Working Inside a Repository 181
- 7.5 Instruction Layers: Global, Project, and Session 182
- 7.6 Writing a Safe Project Instruction File 183
- 7.7 The Core Claude Code Workflow 184
- 7.8 File Reads, Diffs, and Reviewable Edits 185
- 7.9 Shell Commands and Permission Prompts 187
- 7.10 Configuring Safe Command Behavior 187
- 7.11 Slash Commands as Repeatable Workflows 188
- 7.12 Custom Commands for Team Conventions 190
- 7.13 Hooks as Quality Gates 191
- 7.14 Hook Design: Useful, Small, and Observable 192
- 7.15 MCP Integration in Claude Code 193
- 7.16 Context Management in Long Sessions 194
- 7.17 Clear, Compact, Restart, and Handoff Notes 195
- 7.18 Workflow 1: Small Bug Fix 196
- 7.19 Workflow 2: Test-First Change 197
- 7.20 Workflow 3: Code Review and Safety Audit 197
- 7.21 Walkthrough: Adding CLI Argument Validation 198
- 7.22 Where Claude Code Fits Best 200
- 7.23 Where Claude Code Is Not the Best Fit 201
- 7.24 Claude Code Safety Checklist 202
- 7.25 Hands-On Lab: Build a Safe Claude Code Workflow 203
- 7.26 Interview Questions 204
- 7.27 Chapter Summary 206
- 7.28 What Comes Next 208
-
8 Aider Deep Dive 209
- 8.1 Why Aider Deserves Its Own Deep Dive 210
- 8.2 Aider Through the Harness Anatomy 210
- 8.3 Installation and First Run, Without Overfitting to Today's Syntax 212
- 8.4 Aider's Center of Gravity: Repo Maps and Git 213
- 8.5 Working Inside a Git Repository 213
- 8.6 The Repo Map: Structural Context Without Full Files 214
- 8.7 What Repo Maps Are Good At 216
- 8.8 What Repo Maps Cannot Replace 217
- 8.9 Selecting Files for the Conversation 218
- 8.10 Keeping the File Set Small Enough to Reason About 219
- 8.11 Git as a Safety Net 219
- 8.12 Auto-Commit, Diff Review, and Undo 221
- 8.13 Edit Formats and Patch Reliability 222
- 8.14 When Edits Fail to Apply 223
- 8.15 Architect Mode: Planning and Editing as Separate Roles 223
- 8.16 Configuration With .aider.conf.yml 225
- 8.17 Model Selection and Provider Flexibility 226
- 8.18 Local Models With Aider 226
- 8.19 Workflow 1: Small Targeted Edit 227
- 8.20 Workflow 2: Repo-Map-Assisted Refactor 228
- 8.21 Workflow 3: Test-Driven Bug Fix 229
- 8.22 Walkthrough: Adding Type Hints and Docstrings 229
- 8.23 Where Aider Fits Best 231
- 8.24 Where Aider Is Not the Best Fit 232
- 8.25 Aider Safety Checklist 233
- 8.26 Hands-On Lab: Compare Repo Map Editing 234
- 8.27 Interview Questions 235
- 8.28 Chapter Summary 237
- 8.29 What Comes Next 238
-
9 OpenCode, Goose, and Codex 239
- 9.1 Opening Scene: Three Agents, Same Bug, Different Loops 240
- 9.2 Why This Chapter Is Not a Product Ranking 241
- 9.3 The Anatomy Lens Returns 242
- 9.4 OpenCode's Center of Gravity 243
- 9.5 OpenCode Through the Five-Component Anatomy 243
- 9.6 A Conceptual OpenCode Workflow 244
- 9.7 Where OpenCode Fits Well 246
- 9.8 Where OpenCode Is a Poor Fit 246
- 9.9 Goose's Center of Gravity 247
- 9.10 Goose Through the Five-Component Anatomy 248
- 9.11 A Conceptual Goose Workflow 249
- 9.12 Goose as Automation Surface, Not Only Coding Surface 251
- 9.13 Where Goose Fits Well 251
- 9.14 Where Goose Is a Poor Fit 252
- 9.15 Codex's Center of Gravity 253
- 9.16 Codex Through the Five-Component Anatomy 254
- 9.17 A Conceptual Codex Workflow 254
- 9.18 Codex and the Delegated-Task Loop 256
- 9.19 Where Codex Fits Well 258
- 9.20 Where Codex Is a Poor Fit 258
- 9.21 Same Bug, Three Harness Loops 259
- 9.22 Comparison Table: Centers of Gravity 260
- 9.23 Comparison Table: Anatomy Mapping 261
- 9.24 Permission and Tool-Surface Differences 262
- 9.25 Runtime Placement: Local, Remote, and Hybrid 263
- 9.26 Configuration and Reproducibility 263
- 9.27 Reviewability as the Deciding Factor 265
- 9.28 Choosing a Harness by Task Shape 265
- 9.29 A Small Field Guide for Mixed-Tool Teams 266
- 9.30 Safety Checklist 267
- 9.31 Lab 9: Compare Three Harness Loops on One Bounded Change 268
- 9.32 Interview Questions 270
- 9.33 Chapter Summary 272
- 9.34 What Comes Next 273
-
10 Running Harnesses on Local Models 275
- 10.1 Opening Scene: The Same Harness, a Different Model Boundary 276
- 10.2 What ``Local Model'' Means in a Harness 277
- 10.3 Why Local Models Matter for Agentic Coding 278
- 10.4 Why Local Models Do Not Solve Everything 279
- 10.5 The Five-Component Anatomy With a Local Model 279
- 10.6 Local Inference Engines: Conceptual Landscape 281
- 10.7 Model Formats, Quantization, and Hardware in Practical Terms 282
- 10.8 A Conceptual Local Endpoint 283
- 10.9 Local Models and Repository Context 284
- 10.10 Local Models and Tool Use 284
- 10.11 Local Models and Structured Output 286
- 10.12 Local Models and Coding Tasks 287
- 10.13 Task Routing: Local, Hosted, or Hybrid 288
- 10.14 The Hybrid Harness Pattern 290
- 10.15 Privacy Boundaries and Mistaken Assumptions 290
- 10.16 Cost Boundaries and Mistaken Assumptions 291
- 10.17 Latency and Developer Experience 292
- 10.18 Evaluation: The Missing Discipline 293
- 10.19 A Small Local-Model Evaluation Suite 294
- 10.20 Failure Modes of Local Models in Coding Harnesses 295
- 10.21 Designing Prompts for Local Models 296
- 10.22 Context Compression and Local Models 297
- 10.23 Local Models in Claude Code, Aider, OpenCode, Goose, and Codex-Like Workflows 298
- 10.24 Configuration Examples 299
- 10.25 Security Checklist for Local Model Use 300
- 10.26 Same Task, Three Model Placements 302
- 10.27 Decision Table: When to Use Local Models 303
- 10.28 Operational Habits 304
- 10.29 Lab 10: Build a Local-Model Routing Plan 305
- 10.30 Interview Questions 306
- 10.31 Chapter Summary 308
- 10.32 What Comes Next 309
-
11 Workspace Configuration & AGENTS.md Files 310
- 11.1 When the Agent Does Not Know the Project 311
- 11.2 Project Instructions as a Repository Contract 311
- 11.3 Global Instructions vs. Project Instructions 312
- 11.4 Session Prompts Are Not Project Configuration 314
- 11.5 The Anatomy of a Good AGENTS.md 315
- 11.6 Project Purpose and Scope 316
- 11.7 Repository Layout Without Oversharing 317
- 11.8 Build, Test, Lint, and Format Commands 318
- 11.9 Coding Conventions That Agents Can Follow 319
- 11.10 Generated Files and Source-of-Truth Rules 319
- 11.11 Dependency and Package-Management Rules 320
- 11.12 Security Constraints and Secret Boundaries 321
- 11.13 Forbidden, Confirmed, and Safe Operations 322
- 11.14 Completion Checklists 323
- 11.15 What Not to Put in Project Instructions 324
- 11.16 Keeping Instructions Short Enough to Stay Useful 325
- 11.17 Configuration Across Different Harnesses 326
- 11.18 Claude-Style Project Instructions 327
- 11.19 Aider Configuration and File Selection 328
- 11.20 OpenCode, Goose, and Codex Configuration Concepts 329
- 11.21 Local-Model Notes in Project Configuration 330
- 11.22 Before and After: Rewriting a Weak AGENTS.md 331
- 11.23 Walkthrough: Teaching sample_cli Its Own Rules 333
- 11.24 Testing Whether the Harness Follows Instructions 334
- 11.25 Maintaining Project Instructions Over Time 336
- 11.26 Workspace Configuration Checklist 336
- 11.27 Hands-On Lab: Write and Test an AGENTS.md 337
- 11.28 Interview Questions 339
- 11.29 Chapter Summary 341
- 11.30 What Comes Next 341
-
12 Choosing a Harness Per Task 343
- 12.1 The Wrong Harness for the Right Task 343
- 12.2 Harness Selection Is Task Selection 344
- 12.3 The Six-Step Selection Model 345
- 12.4 Step 1: Classify the Task 347
- 12.5 Step 2: Estimate Scope and Context Pressure 348
- 12.6 Step 3: Identify Required Capabilities 349
- 12.7 Step 4: Choose Permission Posture 350
- 12.8 Step 5: Choose Local, Cloud, or Hybrid 351
- 12.9 Step 6: Define Verification Before Starting 351
- 12.10 Task Category 1: Small Bug Fix 353
- 12.11 Task Category 2: Test-First Change 354
- 12.12 Task Category 3: Large Refactor 354
- 12.13 Task Category 4: Greenfield Feature 356
- 12.14 Task Category 5: Documentation Update 356
- 12.15 Task Category 6: Code Review 357
- 12.16 Task Category 7: Security Audit 358
- 12.17 Task Category 8: Dependency Upgrade 359
- 12.18 Task Category 9: CI Repair Loop 360
- 12.19 Task Category 10: Low-Cost Local Batch Work 361
- 12.20 Claude Code Fit 362
- 12.21 Aider Fit 363
- 12.22 OpenCode Fit 364
- 12.23 Goose Fit 364
- 12.24 Codex Fit 365
- 12.25 Local Model Fit 365
- 12.26 Cloud Model Fit 366
- 12.27 Permission Postures by Task Risk 367
- 12.28 Decision Tree: Pick the Harness 369
- 12.29 Scenario Walkthroughs 371
- 12.30 Common Selection Mistakes 373
- 12.31 Hands-On Lab: Build Your Harness Decision Matrix 374
- 12.32 Interview Questions 375
- 12.33 Chapter Summary 377
- 12.34 What Comes Next 377
-
13 Cost, Speed, and Capability (Cloud vs. Local) 379
- 13.1 The Cheapest Model Is Not Always the Cheapest Workflow 380
- 13.2 Why Agentic Loops Cost Differently Than Chat 381
- 13.3 The Unit of Analysis: The Whole Task 381
- 13.4 Cloud Cost Components 382
- 13.5 Input Tokens, Output Tokens, and Cached Tokens 383
- 13.6 Turns, Retries, and Tool-Result Growth 384
- 13.7 Prompt Caching and Its Limits 385
- 13.8 Local Cost Components 386
- 13.9 Hardware Amortization Without Hardware Hype 388
- 13.10 Electricity, Maintenance, and Setup Time 389
- 13.11 Human Time as a Real Cost 389
- 13.12 Speed: Latency, Throughput, and Tool Time 390
- 13.13 Why Round Trips Matter in Agent Loops 391
- 13.14 Capability: Reliability Under Tool Use 392
- 13.15 Structured Output and Tool-Calling Failure Rates 393
- 13.16 Context Length and Capability Degradation 394
- 13.17 Privacy and Data Boundaries 395
- 13.18 Why Local Does Not Automatically Mean Safe 395
- 13.19 Cloud Cost Formula 396
- 13.20 Local Cost Formula 397
- 13.21 Break-Even Formula 397
- 13.22 Retry-Adjusted Cost 399
- 13.23 Human-Time-Adjusted Cost 400
- 13.24 Worked Example 1: Documentation Update 400
- 13.25 Worked Example 2: Focused Bug Fix 401
- 13.26 Worked Example 3: Large Refactor 402
- 13.27 Worked Example 4: Read-Only Security Audit 403
- 13.28 Worked Example 5: Repetitive Local Batch Task 404
- 13.29 Hybrid Routing Strategies 405
- 13.30 Routing by Task Shape, Risk, and Verification 407
- 13.31 Common Cost Mistakes 408
- 13.32 Hands-On Lab: Build a Cost and Capability Worksheet 409
- 13.33 Interview Questions 411
- 13.34 Chapter Summary 413
- 13.35 What Comes Next 414
-
14 Multi-Harness Workflows 415
- 14.1 One Harness Does Not Have to Do Everything 416
- 14.2 What a Multi-Harness Workflow Is 417
- 14.3 When Multi-Harness Workflows Help 417
- 14.4 When They Add Unnecessary Complexity 418
- 14.5 The Phase Model: Plan, Edit, Verify, Audit, Review 419
- 14.6 Handoff Artifacts 420
- 14.7 Context Handoff Without Context Pollution 422
- 14.8 Git as Shared Workflow Memory 423
- 14.9 Keeping Review Boundaries Clean 423
- 14.10 Workflow 1: Plan With Claude Code, Edit With Aider 424
- 14.11 Workflow 2: Read-Only Local Audit, Cloud Fix 426
- 14.12 Workflow 3: Aider Refactor, Claude Code Review 427
- 14.13 Workflow 4: Local Documentation Sweep, Cloud Final Review 428
- 14.14 Workflow 5: CI Repair Loop With Strict Limits 429
- 14.15 Workflow 6: Sandboxed Dependency Upgrade 430
- 14.16 Workflow 7: Shared MCP Validation Tool 431
- 14.17 Designing Safe Handoff Notes 432
- 14.18 Designing Plan Files That Another Harness Can Execute 434
- 14.19 Avoiding Cross-Tool Drift 435
- 14.20 Avoiding Duplicate Work 436
- 14.21 Avoiding Permission Escalation 436
- 14.22 Multi-Harness Security Checklist 438
- 14.23 Walkthrough: Refactoring task_tracker With Two Harnesses 439
- 14.24 Common Multi-Harness Mistakes 443
- 14.25 Hands-On Lab: Design a Two-Harness Workflow 444
- 14.26 Interview Questions 445
- 14.27 Chapter Summary 447
- 14.28 What Comes Next 448
-
15 Build Your Own Minimal Harness 450
- 15.1 Why Build a Harness Yourself? 451
- 15.2 The Minimal Harness Feature Set 451
- 15.3 The Target Architecture 452
- 15.4 What This Teaching Harness Will Not Do 454
- 15.5 Representing Tool Definitions 455
- 15.6 Designing a Stable Tool Result Shape 456
- 15.7 Building the LLM Client Boundary 458
- 15.8 Using a Fake Client for Deterministic Tests 459
- 15.9 Safe Path Resolution 460
- 15.10 Implementing read_file 462
- 15.11 Implementing a Conservative Test Runner 463
- 15.12 Why Generic Shell Access Is Dangerous 464
- 15.13 Implementing the Permission Gate 465
- 15.14 Classifying Commands as Allow, Confirm, or Deny 467
- 15.15 Building the Context for Each Turn 468
- 15.16 Dispatching Tool Calls 469
- 15.17 Re-Injecting Tool Results as Observations 470
- 15.18 Termination Conditions 472
- 15.19 The Main Loop in Code 472
- 15.20 Walkthrough: Fixing parse_count in sample_cli 475
- 15.21 Testing the Harness 476
- 15.22 What Commercial Harnesses Add 479
- 15.23 Where This Minimal Harness Is Unsafe 481
- 15.24 How to Extend It Safely 481
- 15.25 Hands-On Lab: Build the Minimal Harness 482
- 15.26 Interview Questions 484
- 15.27 Chapter Summary 486
- 15.28 What Comes Next 486
-
16 Staying Current in a Fast-Moving Ecosystem 488
- 16.1 The Ecosystem Will Keep Moving 489
- 16.2 What Changes Fast and What Changes Slowly 489
- 16.3 Durable Concepts From This Book 490
- 16.4 Version-Sensitive Surfaces 491
- 16.5 Release Notes Are Not Enough 493
- 16.6 The Harness Update Checklist 494
- 16.7 Evaluating a New Harness 496
- 16.8 Evaluating a New Model 498
- 16.9 Evaluating Local-Model Improvements 499
- 16.10 Evaluating MCP and Tooling Changes 501
- 16.11 Evaluating Permission and Sandbox Changes 502
- 16.12 Maintaining Workspace Instructions 503
- 16.13 Maintaining Cost and Routing Rules 504
- 16.14 Maintaining Benchmark Tasks 505
- 16.15 Building a Small Evaluation Suite 507
- 16.16 Upgrade, Wait, or Roll Back 507
- 16.17 Solo Developer Workflow 509
- 16.18 Team Workflow 510
- 16.19 Security Review Cadence 511
- 16.20 Decision Records for Tooling Changes 512
- 16.21 Common Ways Teams Drift 513
- 16.22 A Practical Adoption Scorecard 515
- 16.23 Hands-On Lab: Evaluate a Harness Update 517
- 16.24 Interview Questions 518
- 16.25 Final Synthesis 520
Agentic Coding Harnesses, Compared & Explained
Master Claude Code, Aider, OpenCode, Goose, and Codex — Then Run Them Locally
A vendor-neutral, mechanism-level field guide to operating and extending agentic coding harnesses (538 manuscript pages).
Minimum price
$19.99
$29.99
You pay
Author earns
PDF
About
About the Book
Agentic Coding Harnesses, Compared & Explained is a practical, command-driven guide for developers and DevOps engineers who want to graduate from chat assistants to autonomous coding agents. It explains the mechanics of the agent loop, compares the architectures of the five major harnesses, details Model Context Protocol (MCP) tool integration, context management, safety sandboxing, local LLM routing, and concludes with a Python capstone build.
Author
About the Author
Yohan Rodriguez
Yohan is a Senior Full-Stack Software Engineer with extensive experience delivering scalable, end-to-end software solutions across web, enterprise, and cloud-based environments. He specializes in architecting robust platforms, modernizing legacy systems, driving cloud transformation efforts, and building integration-heavy applications that support critical business workflows. He is recognized for translating complex requirements into reliable, maintainable, and high-value solutions across industries such as insurance, cybersecurity, and professional services.
Known for combining strong technical execution with a practical business mindset, he has contributed to projects from concept and design through production delivery and long-term support. His experience includes collaborating with cross-functional teams, improving development workflows, solving complex technical challenges, and helping organizations deliver dependable software products that adapt to changing business needs. He brings a balanced approach to engineering that values quality, efficiency, and continuous improvement.
Contents
Table of Contents
Get the free sample chapters
Click the buttons to get the free sample in PDF or EPUB, or read the sample online here
The Leanpub 60 Day 100% Happiness Guarantee
Within 60 days of purchase you can get a 100% refund on any Leanpub purchase, in two clicks.
See full terms...
Earn $8 on a $10 Purchase, and $16 on a $20 Purchase
We pay 80% royalties on purchases of $7.99 or more, and 80% royalties minus a 50 cent flat fee on purchases between $0.99 and $7.98. You earn $8 on a $10 sale, and $16 on a $20 sale. So, if we sell 5000 non-refunded copies of your book for $20, you'll earn $80,000.
(Yes, some authors have already earned much more than that on Leanpub.)
In fact, authors have earned over $15 million writing, publishing and selling on Leanpub.
Learn more about writing on Leanpub
Free Updates. DRM Free.
If you buy a Leanpub book, you get free updates for as long as the author updates the book! Many authors use Leanpub to publish their books in-progress, while they are writing them. All readers get free updates, regardless of when they bought the book or how much they paid (including free).
Most Leanpub books are available in PDF (for computers) and EPUB (for phones, tablets and Kindle). The formats that a book includes are shown at the top right corner of this page.
Finally, Leanpub books don't have any DRM copy-protection nonsense, so you can easily read them on any supported device.
Learn more about Leanpub's ebook formats and where to read them
Write and Publish on Leanpub
You can use Leanpub to easily write, publish and sell in-progress and completed ebooks and online courses!
Leanpub is a powerful platform for serious authors, combining a simple, elegant writing and publishing workflow with a store focused on selling in-progress ebooks.
Leanpub is a magical typewriter for authors: just write in plain text, and to publish your ebook, just click a button. (Or, if you are producing your ebook your own way, you can even upload your own PDF and/or EPUB files and then publish with one click!) It really is that easy.