Sources
1,040 sources behind the study.
Every source the research read, cited or archived: documentation, papers, essays, threads and code. Pages from the same site are grouped together, largest group first.
The research stored a copy of each source with its address, retrieval time and a content hash, so that every claim could be checked against the text as it was read. The links below go to the live pages, which may have changed since.
Anthropic: Claude Code documentation 100
The Claude Code docs pages the research relied on, as they stood in September and October 2026.
code.claude.com 93
- Admin
- Advisor
- Agent teams
- Agent view
- Agents
- Analytics
- Artifacts
- Auto mode config
- Best practices
- Changelog
- Channels
- Channels reference
- Checkpointing
- Chrome
- Claude code on the web
- Claude security
- Cli reference
- Code review
- Commands
- Common workflows
- Computer use
- Context window
- Costs
- Cross session messaging
- Data usage
- Debug your config
- Deep links
- Desktop scheduled tasks
- Discover plugins
- Docs
- Env vars
- Errors
- Events
- Fast mode
- Features overview
- Github actions
- Glossary
- Goal
- Headless
- Hooks
- Hooks
- Hooks guide
- How claude code works
- Index
- Interactive mode
- Keybindings
- Large codebases
- Managed settings
- Mcp
- Memory
- Model config
- Monitoring usage
- Network config
- Output styles
- Overview
- Overview
- Permission modes
- Permissions
- Plugin dependencies
- Plugin evals
- Plugin hints
- Plugin marketplaces
- Plugin relevance
- Plugins
- Plugins reference
- Prompt caching
- Prompt library
- Quickstart
- Reference
- Remote control
- Routines
- Sandbox environments
- Sandboxing
- Scheduled tasks
- Security
- Security guidance
- Security guidance
- Sessions
- Settings
- Settings example
- Settings reference
- Setup
- Skills
- Statusline
- Sub agents
- Subagents
- Terminal config
- Tool search
- Tools reference
- Troubleshooting
- Ultrareview
- Workflows
- Worktrees
claude.dev 7
- Automating eval design and hillclimbing with Claude / claude.dev Blog 2026
- Building with Claude Sonnet 5.5 / claude.dev Blog 2026
- claude.dev Blog / technical writing for people building with Claude 2026
- Getting started with Claude Code mods / claude.dev Blog 2026
- Getting the most out of Opus 5.5 in Claude and Claude Code / claude.dev Blog 2026
- How we made claude.ai 3x faster in two weeks / claude.dev Blog 2026
- Using Claude Code: Spending your effort / claude.dev Blog 2026
Anthropic: model and API documentation 53
Prompting guides, model pages, pricing and platform docs.
platform.claude.com 50
- Agent sdk
- Choosing a model
- Claude Fable 5
- Claude Fable 5 system prompts
- Claude Fable 5.1
- Claude Fable 5.1 system prompts
- Claude Mythos 5.1
- Claude Opus 4.7
- Claude Opus 5
- Claude opus 5 5
- Claude Opus 5 system prompts
- Claude Opus 5.5
- Claude Sonnet 5
- Claude sonnet 5 5
- Coding prompting for frontend aesthetics
- Cookbook
- Define success criteria and build evaluations
- Docs
- Dreams
- Effort
- Migrating to Claude Fable 5.1 and Claude Mythos 5.1
- Migrating to Claude Mythos 5 and Claude Fable 5
- Migrating to Claude Opus 5
- Migrating to Claude Opus 5.5
- Migrating to Claude Sonnet 5
- Migration guide
- Migration guide
- Migration guides
- Model deprecations
- Models overview
- Models overview
- Overview
- Overview
- Pricing
- Prompt engineering overview
- Prompting best practices
- Prompting Claude Fable 5
- Prompting Claude Fable 5.1
- Prompting Claude Opus 4.8
- Prompting Claude Opus 5
- Prompting Claude Opus 5.5
- Prompting Claude Sonnet 5
- Prompting claude sonnet 5 5
- Refusals and fallback
- Steering thinking
- Tool search tool
- What's new in Claude Fable 5.1
- What's new in Claude Opus 5
- What's new in Claude Opus 5.5
- Whats new sonnet 5 5
Anthropic: announcements, engineering and research 46
Model announcements, engineering posts and system cards.
anthropic.com 28
- A statistical approach to model evaluations 2023
- An update on recent Claude Code quality reports 2026
- Best practices for Claude Code - Claude Code Docs 2026
- Building a C compiler with a team of parallel Claudes 2026
- Building Effective AI Agents 2024
- Claude Fable 5.1 & Claude Mythos 5.1 System Card
- Claude sonnet 5 5 system card
- Code execution with MCP: building more efficient AI agents 2025
- Demystifying evals for AI agents 2026
- Effective context engineering for AI agents 2025
- Effective harnesses for long-running agents 2025
- Equipping agents for the real world with Agent Skills 2025
- Eval awareness in Claude Opus 4.6’s BrowseComp performance 2026
- Finance agents
- Harness design for long-running application development 2026
- How we built Claude Code auto mode: a safer way to skip permissions 2026
- How we built our multi-agent research system 2025
- How we contain Claude across products 2026
- Introducing advanced tool use on the Claude Developer Platform 2025
- Introducing Claude Fable 5.1 and Claude Mythos 5.1 2023
- Introducing Claude Opus 5 2023
- Introducing Claude Opus 5.5 2026
- Introducing Claude Sonnet 5.5 2026
- Making Claude Code more secure and autonomous with sandboxing 2025
- Newsroom 2026
- Patterns and problems in multiagent systems 2023
- Scaling Managed Agents: Decoupling the brain from the hands 2026
- Writing effective tools for AI agents—using AI agents 2025
www-cdn.anthropic.com 9
claude.com 8
- Blog | Claude by Anthropic 2026
- Claude Code effort level and model selection | Claude | Claude by Anthropic 2026
- Coding sessions are longer and use more context. Claude Opus 5.5 is built with that in mind. | Claude by Anthropic 2026
- Customize Claude Code with mods in TypeScript | Claude by Anthropic 2026
- Help
- Pricing
- The new rules of context engineering for Claude 5 generation models | Claude by Anthropic 2026
- What a task costs on Opus 5.5 | Claude by Anthropic 2026
assets.anthropic.com 1
Papers 46
Research papers on agents, evaluation, prompting and LLM judges.
arxiv.org 45
- Agent READMEs: An Empirical Study of Context Files for Agentic Coding 2025
- AgentCoder 2023
- Arbiter: Detecting Interference in LLM Agent System Prompts 2026
- Architectural Design Decisions in AI Agent Harnesses 2026
- Beyond Consensus: Mitigating the Agreeableness Bias in LLM Judge Evaluations 2025
- Can AI agents conduct open-ended AI research? Early evidence from two case studies 2026
- Checklists Are Better Than Reward Models For Aligning Language Models 2025
- Code Review Agent Benchmark 2026
- Context Matters: Evaluating Context Strategies for Automated ADR Generation Using LLMs 2026
- ContextWeaver: Selective and Dependency-Structured Memory Construction for LLM Agents 2026
- Correlated Errors in Large Language Models 2025
- Cross-Context Review: Improving LLM Output Quality by Separating Production and Review Sessions 2026
- Diagnosing and Repairing Persona Collapse in LLM Advice 2026
- Do Context Files Help Coding Agents? A Two-Agent Ablation Study on Real Repositories 2026
- Evaluating AGENTS.md: Are Repository-Level Context Files Helpful for Coding Agents? 2026
- EvilGenie: A Reward Hacking Benchmark 2025
- Expert Personas Improve LLM Alignment but Damage Accuracy: Bootstrapping Intent-Based Persona Routing with PRISM 2026
- GEPA: Reflective Prompt Evolution Can Outperform Reinforcement Learning 2025
- Git Context Controller: Manage the Context of LLM-based Agents like Git 2025
- How Many Instructions Can LLMs Follow at Once? 2025
- Instruction Adherence in Coding Agent Configuration Files: A Factorial Study of Four File-Structure Variables 2026
- Just Ask: Curious Code Agents Reveal System Prompts in Frontier LLMs 2026
- Large Language Models Cannot Self-Correct Reasoning Yet 2023
- LLM Evaluators Recognize and Favor Their Own Generations 2024
- Lore: Repurposing Git Commit Messages as a Structured Knowledge Protocol for AI Coding Agents 2026
- Measure Before You Manage: Evaluating Agent Working Memory in Coding Agents 2026
- Measuring AI Ability to Complete Long Software Tasks 2025
- More Rounds, More Noise: Why Multi-Turn Review Fails to Improve Cross-Context Verification 2026
- MT-Bench / LLM-as-a-judge 2023
- Nine Judges, Two Effective Votes: Correlated Errors Undermine LLM Evaluation Panels 2026
- Preference Leakage: A Contamination Problem in LLM-as-a-judge 2025
- Prompting Science Report 4: Playing Pretend: Expert Personas Don't Improve Factual Accuracy 2025
- Reward Hacking Benchmark: Measuring Exploits in LLM Agents with Tool Use 2026
- Rubrics as Rewards: Reinforcement Learning Beyond Verifiable Domains 2025
- Self-Preference Bias in LLM-as-a-Judge 2024
- SpecBench: Measuring Reward Hacking in Long-Horizon Coding Agents 2026
- SWE-MeM: Learning Adaptive Memory Management for Long-Horizon Coding Agents 2026
- Test vs Mutant: Adversarial LLM Agents for Robust Unit Test Generation 2026
- The SWE-Bench Illusion: When State-of-the-Art LLMs Remember Instead of Reason 2025
- TICK 2024
- Tulu 3: Pushing Frontiers in Open Language Model Post-Training 2024
- When "A Helpful Assistant" Is Not Really Helpful: Personas in System Prompts Do Not Improve Performances of Large Language Models 2023
- When Does Persona Prompting Actually Help? A Retrieval and Metric Analysis of Expert Role Injection in LLMs 2026
- When Instructions Multiply: Measuring and Estimating LLM Capabilities of Multiple Instructions Following 2025
- Who Validates the Validators? Aligning LLM-Assisted Evaluation of LLM Outputs with Human Preferences 2024
Discussion threads and posts 42
Threads and posts by Claude Code's developers and practitioners.
x.com 28
- Chollet: 'Sufficiently advanced agentic coding is essentially machine learning' (Feb 19 2026) 2026
- Chollet: test-time training (Aug 13 2026) 2026
- Chollet: two 'patches' to deep learning (Aug 2 2026) 2026
- Designing loops with Fable 5 (Lance Martin, Anthropic, Jun 9 2026)
- Fowler: announcing Böckeler's 'Context Engineering for Coding Agents' (Feb 5 2026) 2026
- Huryn: 'Fable 5.1 dominated the Bug Hunt Bench' (Sep 1 2026) 2026
- Huryn: Opus 5 on the 45-bug bench v1 (Jul 24 2026) 2026
- Karpathy: 'Jagged Intelligence' (Jul 24 2024) 2024
- Karpathy: AI-assisted coding rhythm / 'tight leash' (Apr 24 2025) 2025
- Klaassen: 'Opus 5 is out now! And it broke Compound Engineering' (Jul 24 2026) 2026
- User's index of the ten X captures 2026
- X thread by @bcherny (2007179832300581177)
- X thread by @claudeai (2102435538120691886)
- X thread by @HamelHusain (2102528342859915487)
- X thread by @karpathy (2017386421712261612)
- X thread by @karpathy (2026731645169185220)
- X thread by @omarsar0 (2082550450373451938)
- X thread by @PawelHuryn (2102479269335101504)
- x-catwu-2102437713781944397
- x-lennysan-2039845666680176703
- x-omarsar0-2040099881008652634
- x.com/bcherny/status/2017742741636321619
- x.com/cheatyyyy/status/2080712740965106121 2026
- x.com/chenchengpro/status/2081339012632215684 2026
- x.com/MyWestLord/status/2082855525956415987 2026
- x.com/PawelHuryn/status/2079700261581271487 2026
- x.com/trq212/article/2080710971228918066 2026
- x.com/trq212/status/2080710971228918066 2026
news.ycombinator.com 7
reddit.com 4
linkedin.com 2
threadreaderapp.com 1
Other vendors' documentation 19
Documentation from other model and tool vendors, used for comparison.
developers.openai.com 9
artificialanalysis.ai 4
- Articles | Artificial Analysis 2026
- Claude Opus 5.5 takes the top spot on the Artificial Analysis Intelligence Index 2026
- Claude Sonnet 5.5 (max with fallback) - Intelligence, Performance & Price Analysis | Artificial Analysis 2026
- Claude Sonnet 5.5 reaches #2 on the Artificial Analysis Intelligence Index 2026
bun.com 2
- Blog | Bun 2026
- Rewriting Bun in Rust 2026
cursor.com 2
cookbook.openai.com 1
swebench.com 1
Code repositories 106
Repositories cited or archived during the research, beyond the 46 reviewed in depth. Files from one repository are listed once.
gist.github.com 14
github.com/anthropics 7
github.com/nilenso 3
github.com/obra 3
github.com/elder-plinius 2
github.com/garrytan 2
github.com/humanlayer 2
github.com/microsoft 2
github.com/phuryn 2
github.com/spences10 2
github.com/A3S-Lab 1
github.com/All-Hands-AI 1
github.com/AutomationPanda 1
github.com/BecreativeLTD 1
github.com/CherryHQ 1
github.com/DMontgomery40 1
github.com/Eversmile12 1
github.com/EveryInc 1
github.com/Exhen 1
github.com/MiniMax-AI 1
github.com/NabilAziz99 1
github.com/Piebald-AI 1
github.com/Sudip-Shrestha0x0 1
github.com/TrevorS 1
github.com/ai-evals-course 1
github.com/asgeirtj 1
github.com/bgauryy 1
github.com/birgitta410 1
github.com/bmad-code-org 1
github.com/bukzor 1
github.com/chenditc 1
github.com/chouzz 1
github.com/cline 1
github.com/dmmulroy 1
github.com/fitchmultz 1
github.com/ghuntley 1
github.com/google-gemini 1
github.com/habit-hooks 1
github.com/hankimis 1
github.com/huybuidac 1
github.com/imkunal007219 1
github.com/judell 1
github.com/jujumilk3 1
github.com/karpathy 1
github.com/kekmodel 1
github.com/kenhuangus 1
github.com/laurigates 1
github.com/lexler 1
github.com/lucas-flatwhite 1
github.com/mattpocock 1
github.com/minzique 1
github.com/netease-lcap 1
github.com/nklisch 1
github.com/opslane 1
github.com/paulnsorensen 1
github.com/riba2534 1
github.com/rileyhilliard 1
github.com/seifghazi 1
github.com/shanraisshan 1
github.com/shimo4228 1
github.com/shivamagarwal7 1
github.com/skills-directory 1
github.com/snapxmin 1
github.com/song-chaoyang 1
github.com/sqlew-io 1
github.com/sst 1
github.com/stlwolf 1
github.com/sunitghub 1
github.com/theangrygiraffe 1
github.com/uwuclxdy 1
github.com/wtfwhs 1
github.com/x1xhlol 1
github.com/ysxdyt-001 1
github.com/zai-org 1
Essays, blogs and newsletters 628
Practitioner writing on coding agents, grouped by site, largest first.
lexler.github.io 98
- Active Partner
- Advisor Strategy
- AI First Responder
- AI Isolation
- AI Slop
- Answer Injection
- Approved Logs
- Approved Scenarios
- Augmented Coding Patterns 2020
- Background Agent
- Big Chunk Working
- Black Box AI
- Borrow Behaviors
- Canary in the Code Mine
- Cannot Learn
- Cast Wide
- Chain of Small Steps
- Check Alignment
- Chunking
- Coerce to Interface
- Cognitive Overload
- Complete Catalog 2020
- Compliance Bias
- Constrained Tests
- Context Management
- Context Markers
- Context Rot
- Contextual Prompts
- Contributors
- Decision Guards
- Degrades Under Complexity
- Diagrams as Code
- Distracted Agent
- Excess Verbosity
- External Context
- Extract Knowledge
- Feedback Flip
- Feedback Loop
- Flying Blind
- Focused Agent
- Ground Rules
- Habit Hooks
- Hallucinations
- Happy to Delete
- Hooks
- JIT Docs
- Keeping up
- Knowledge Checkpoint
- Knowledge Composition
- Knowledge Document
- Learning Loop
- Limited Context Window
- Limited Focus
- Memory Grooming
- Mind Dump
- Negative Bleedthrough 2020
- Noise Cancellation
- Non-Determinism
- Obedient Contractor
- Obsess Over Rules
- Offload Deterministic
- Orchestrator
- Overnight Batch
- Parallel Implementations
- Patterns for Coding with AI 2020
- Perfect Recall Fallacy
- Phased Delivery
- Playgrounds
- Point the Target
- Polyglot AI
- Poor Person's Skill
- Reference Docs
- Refinement Loop
- Reminders
- Reverse Direction
- ROSE Feedback
- Selective Hearing
- Semantic Anchors
- Semantic Zoom
- Shared Canvas
- Show the Agent, Let it Repeat/Automate
- Silent Misalignment
- Slice for Review
- Smart Plan, Cheap Execution
- Softest Prototype
- Solution Fixation
- Speak to Me
- Sunk Cost
- Surface Change Attractors
- Take All Paths
- Tell Me a Lie
- Text Native
- Unvalidated Leaps
- Yak Shave Delegation
- You Do It
- ⚠️Anti-Patterns
- ⛰️Obstacles
- 🧩Patterns
kenhuangus.substack.com 57
- An Implementation Checklist to Claude Code in Large Codebases 2026
- Anthropic’s Fable 5.1 guide reads like a manual for agent product design, not a collection of prompt tricks (Ken Huang, 2026-09-02) 2026
- Chapter 10: Production Deployment Patterns (Claude Code vs. Hermes Agent) 2026
- Chapter 11: Hook / Event-Driven Automation (Claude Code vs. Hermes Agent) 2026
- Chapter 12: The Skill System Pattern (Claude Code vs. Hermes Agent) 2026
- Chapter 13: MCP Integration — Connecting Agents to the World (Claude Code vs. Hermes Agent) 2026
- Chapter 14: Model Routing and Provider Abstraction (Claude Code vs. Hermes Agent) 2026
- Chapter 15: Structured Output and Schema-Constrained Generation (Claude Code vs. Hermes Agent) 2026
- Chapter 1: Hermes Agent: Cost & Token-Usage Accounting (Claude Code vs. Hermes Agent) 2026
- Chapter 1: The Harness Paradigm (Claude Code vs. Hermes Agent) 2026
- Chapter 2: Cancellation & Abort Propagation (Claude Code vs. Hermes Agent) 2026
- Chapter 2: Tool Architecture and the Tool Contract (Claude Code vs. Hermes Agent) 2026
- Chapter 3: The Query / Agent Loop (Claude Code vs. Hermes Agent) 2026
- Chapter 3: The Slash Command System (Claude Code vs. Hermes Agent) 2026
- Chapter 4: Permission Systems and Safety Guardrails (Claude Code vs. Hermes Agent) 2026
- Chapter 4: Working Directory & File-Path Resolution (Claude Code vs. Hermes Agent) 2026
- Chapter 5: Tool Orchestration and Execution (Claude Code vs. Hermes Agent) 2026
- Chapter 5: Trajectory Compression and Replay (Claude Code vs. Hermes Agent) 2026
- Chapter 6: Context Management at Scale (Claude Code vs. Hermes Agent) 2026
- Chapter 6: Terminal UI Architecture (Claude Code vs. Hermes Agent) 2026
- Chapter 7: Multi-Agent Coordination (Claude Code vs. Hermes Agent) 2026
- Chapter 8: Memory Systems and State Persistence (Claude Code vs. Hermes Agent) 2026
- Chapter 9: Observability and Debugging (Claude Code vs. Hermes Agent) 2026
- Claude Agents Can Now Dream: How AI Engineers Should Use Anthropic’s New Agent Features Without Creating New Attack Paths 2026
- Claude Code Harness Pattern 10: Production Deployment Patterns 2026
- Claude Code Harness Pattern 2: Tool Architecture and the Tool Contract 2026
- Claude Code Harness Pattern 3: The Query Engine — Orchestrating AI Conversations 2026
- Claude Code Harness Pattern 4: Permission Systems and Safety Guardrails 2026
- Claude Code Harness Pattern 8: Memory Systems and State Persistence 2026
- Claude Code Harness Pattern 9: Observability and Debugging 2026
- CLAUDE CODE ORCHESTRATION 2026
- Claude Code Pattern 5: Tool Orchestration and Execution 2026
- Claude Code Pattern 6: Context Management at Scale 2026
- Claude Code Pattern 7: Multi-Agent Coordination 2026
- Claude Fable (2 more days), Part 9: Don't Pay Twice for a Fallback (, 2026-07-10) 2026
- Claude Fable (3 more days), Part 6: Ten Fable 5-Native Agent Skills (, 2026-07-09) 2026
- Claude Fable 5 (Part 1): What Changed, and How to Stop Prompting It Like Opus (Ken Huang, 2026-07-04) 2026
- Claude Fable 5(last day), Part 10 and Book Announcement (Ken Huang, 2026-07-11) 2026
- Claude Fable 5, Part 2: Loop Engineering (Ken Huang, 2026-07-05) 2026
- Claude Fable 5, Part 3: Memory Engineering (Ken Huang, 2026-07-07) 2026
- Claude Fable 5, Part 4: Ten Weekend Security Projects (Ken Huang, 2026-07-08) 2026
- Claude Fable 5, Part 7: Moving a Production Fleet Off Opus 4.8 (, 2026-07-09) 2026
- Claude Fable 5, Part 8: The Refusal Your Error Handler Can't See (, 2026-07-10) 2026
- CLAUDE.md Modernization and Prefix-Stable KV Caching for Claude Code (Ken Huang, 2026-09-13) 2026
- Coding Agent Security: Lessons from Claude Code, Cowork, Codex, and Copilot in the Wild 2026
- Comparison Report: GPT-5.6 Sol, Fable5, and the Muse Spark 1.1, Grok 4.5, 2026
- Designing Hermes Agent from Scratch: A Systems Deep Dive 2026
- Found from Claude Code: Chapter 1: The Harness Paradigm 2026
- From Software Engineering to Harness Engineering: What OpenAI's Codex Actually Builds 2026
- Good News! Claude Fable extended to July 12th, We will release 6 more patterns of how to use it. (Ken Huang, 2026-07-08) 2026
- GPT-5.6 Is More Capable, More Autonomous, and More Willing to Overstep You 2026
- GPT-6 Astra plus Kimi K3 Swarm: route depth and width before you pay frontier rates 2026
- Harness Engineering as the Umbrella Discipline 2026
- I Ran Qwen3.5-35B-A3B Locally with Cline Code Agent For Free, Forever 2026
- I timed Grok Build TUI against Claude Code on the same task. One was 6x faster. 2026
- The Claude Code Leak: 10 Agentic AI Harness Patterns That Change Everything 2026
- The Harness Advantage in Autonomous Red Teaming: Why Frontier LLMs Alone Fail Offensive Security and How RidgeGen Solves the Alignment Dilemma 2026
simonwillison.net 51
- 2026 in LLMs (so far) 2026
- A Fireside Chat with Cat and Thariq from the Claude Code team 2026
- A quote from Anthropic Frontier Red Team 2026
- A quote from Boris Cherny 2026
- Adding a new content type to my blog-to-newsletter tool - Agentic Engineering Patterns 2026
- Agentic manual testing - Agentic Engineering Patterns 2024
- AI should help us produce better code - Agentic Engineering Patterns 2024
- Anti-patterns: things to avoid - Agentic Engineering Patterns 2024
- Claude Cowork and chat are now one Claude 2026
- Claude Opus 5.5, GPT-6 Sol, GPT-6 Luna, and a new price war 2026
- Claude Skills are awesome, maybe a bigger deal than MCP 2025
- Claude Sonnet 5.5 2026
- claude-trace 2025
- Code research projects with async coding agents like Claude Code and Codex 2025
- Comment: GPT 6.1 Sol: Near-Astra intelligence for a fifth of the price 2026
- Designing agentic loops 2025
- Embracing the parallel coding agent lifestyle 2025
- Everything 2026
- First run the tests - Agentic Engineering Patterns 2024
- FLUX.2-klein-4B Pure C Implementation 2026
- GIF optimization tool using WebAssembly and Gifsicle - Agentic Engineering Patterns 2024
- GPT-5.5 prompting guide 2026
- Have your agent record video demos of its work with shot-scraper video 2026
- Hoard things you know how to do - Agentic Engineering Patterns 2024
- How coding agents work - Agentic Engineering Patterns 2024
- Interactive explanations - Agentic Engineering Patterns 2024
- Linear walkthroughs - Agentic Engineering Patterns 2024
- Living dangerously with Claude 2025
- More than just code review 2026
- Note on 24th September 2026 2026
- OpenAI DevDay 2026 live blog 2026
- Prompts I use - Agentic Engineering Patterns 2024
- Red/green TDD - Agentic Engineering Patterns 2024
- Release: sqlite-utils 4.0 2026
- Release: sqlite-utils 4.0rc3 2026
- Release: sqlite-utils 4.0rc4 2026
- Simon Willison on claude 2026
- Simon Willison’s Weblog 2024
- sqlite-utils 4.0, now with database schema migrations 2026
- sqlite-utils 4.0a1 has several (minor) backwards incompatible changes 2025
- sqlite-utils 4.0rc1 adds migrations and nested transactions 2026
- sqlite-utils 4.0rc2, mostly written by Claude Fable (for about $149.25) 2026
- sqlite-utils 4.0rc2, mostly written by Claude Fable (for about $149.25) 2026
- Subagents - Agentic Engineering Patterns 2024
- The cognitive impact of coding agents 2026
- Using Git with coding agents - Agentic Engineering Patterns 2024
- Using llms for code
- Vibe engineering 2025
- We’re going to need default hard budget caps on pretty much everything 2026
- What is agentic engineering? - Agentic Engineering Patterns 2024
- Writing code is cheap now - Agentic Engineering Patterns 2024
martinfowler.com 30
- An Accidental Blackboard 2026
- Anchoring AI to a reference application 2025
- Assessing internal quality while coding with an agent 2026
- bliki: Sensible Default 2026
- Böckeler: Context Engineering for Coding Agents (user copy) 2026
- Context Anchoring 2026
- DSLs Enable Reliable Use of LLMs 2026
- Experiences with local models for coding 2026
- Exploring Generative AI 2025
- Feed.atom 2026
- Fragments: April 2 2026
- Fragments: August 18 2026
- Fragments: July 13 2026
- Fragments: September 1 2026
- Fragments: September 24 2026
- Fragments: September 29 2026
- Fragments: September 8 2026
- Harness Engineering 2026
- Harness engineering for coding agent users 2026
- Humans and Agents in Software Engineering Loops 2026
- I don't like LLMs 2026
- Maintainability sensors for coding agents 2026
- Maybe We Shouldn't Be Reviewing All This Code 2026
- Partner with the AI, throw away the code 2025
- TDD inside the agent loop - theater or actual value? 2026
- The Conductor Developer 2026
- The Economic Benefit of Refactoring 2026
- The Orchestrator's Tax 2026
- To vibe or not to vibe 2025
- Understanding Spec-Driven-Development: Kiro, spec-kit, and Tessl 2025
paddo.dev 28
- A Dial Worth Turning: Claude Opus 5's Prose, and the Style Guide Anthropic Wrote Against Its Own Model 2026
- Bring Your Own Model: Claude Code on a ChatGPT Subscription, and the Lock-In That Isn't 2026
- Careful, Not Thorough: Claude Opus 5.5 vs GPT-6 Sol on Real Code 2026
- Emergent Minds | paddo.dev 2026
- GPT-6 Astra Is On Every Plan: What It Costs, What It's Good At, and Which Effort Level to Use 2026
- Inside the Harness: Building a Claude Code Mod 2026
- Opening the Folder Was the Exploit: GitSpawn, Seven Coding Agents, and a Bug VS Code Fixed in 2021 2026
- Papering Faster: AI Coding Agents and the Root-Cause Fix Nobody Ships 2026
- The Bar Moved, Not the Speed: Claude Opus 5.5 in Practice 2026
- The Boring Stack Is Cheap Again: Firebase to Postgres, Better Auth and SvelteKit 2026
- The Bulk Discount Is Gone: OpenAI's Pro 500, the Pro 200 Cut and GPT-6.1 Sol on Real Work 2026
- The Cage Is in the Table Now: Claude Fable 5.1, Mythos 5.1, and the Five Points Safety Costs 2026
- The Careful One Got Cheap: Claude Sonnet 5.5 vs Opus 5.5 and GPT-6 Sol 2026
- The Default Was Right: Opus 5.5, Sonnet 5.5 and GPT-6 Sol Effort Levels on Real Work 2026
- The Dial Next to the Meter: Stretching Claude Fable 5 Credits with Effort Levels 2026
- The Expensive Middle: Claude Opus 4.8 vs Sonnet 5 2026
- The Model Knows It's a Test: Anthropic's IPO Filing Makes Evaluation Awareness a Risk Factor 2026
- The Permission Tier: Claude Fable 5 Comes Back Changed 2026
- The Raise That Is a Cut: Claude Code Weekly Limits, Up 25% and Down 17% 2026
- The Rollback Is the Product: Feature Flags, Canaries, and Config in the Agent Era 2026
- The Thirty-Cent Judge: TypeSafe's Jev on a Real Product-Matching Queue 2026
- The Verifier Got Cheap: What TypeScript 7 Actually Changes for Coding Agents 2026
- The Viral CLAUDE.md and Mine 2026
- The Voice Drifts Back: AI Slop, Concise Mode, and Simplified Technical English 2026
- The Warning Was in the Manual: GPT-5.6 Sol and the Deleted Database 2026
- Three Seats, One Mac: Running Multiple Claude Code Accounts With Shared Config 2026
- Two Meters: Claude Fable 5.1 Is Cheaper on the API and Hungrier on Max, and What to Turn Down 2026
- Your CLAUDE.md Is Now Technical Debt: What Claude Opus 5 Wants You to Delete 2026
dbreunig.com 24
- 10 Lessons for Agentic Coding 2026
- Don’t Fight the Weights 2025
- Fable & The End of the Free Lunch 2026
- Feed.xml 2026
- Harnesses are Situated Agents 2026
- How Claude Code Builds a System Prompt 2026
- How Long Contexts Fail 2025
- How Long Contexts Fail 2025
- How System Prompts Define Agent Behavior 2026
- How to Fix Your Context 2025
- Learnings from a No-Code Library: Keeping the Spec Driven Development Triangle in Sync 2026
- Let the Model Write the Prompt 2025
- Manage Your Agent’s Loadout with Dr. Skill 2026
- Prompts vs. Context 2025
- The 2nd Phase of Agentic Development 2026
- The Cathedral, the Bazaar, and the Winchester Mystery House 2026
- The Cost of Overfitting the Harness 2026
- The Problem is Prompt Debt 2026
- Two Beliefs About Coding Agents 2026
- We’re Talking About Terms of Use, But the Issue is Embedded Judgment 2026
- What Do Humans Need From Docs? 2026
- What We Can Learn from Claude’s Fable 5.1 System Prompt 2026
- Who Taught the Models to Do That? 2026
- Why is Claude an Electron App? 2026
blog.fsck.com 21
- A quick trip to the uncanny valley 2026
- A small code review prompt hack 2025
- Automating agentic development 2025
- Blog.fsck.com
- Claude Code skills not triggering? It might not see them. 2025
- Fixing Claude Code's amnesia 2025
- Greenfield and Iterative Development 2026
- I asked muse to tell me about updates to its skills 2026
- I vibe-coded a C compiler that can build SQLite 2026
- My favorite adversarial review prompt 2026
- Parts of Gemini CLI's system prompt look eerily familiar 2025
- Rules and Gates 2026
- SF: A Birds of a Feather on Agentic Engineering 2026
- Skills for Claude! 2025
- Some new agentic patterns 2026
- Superpowers 5.1.0 2026
- Superpowers 6 2026
- Superpowers 6.4 2026
- Superpowers: How I'm using coding agents in October 2025 2025
- That time it tried to delete all my tests 2026
- The Therapist Pattern 2026
blog.nilenso.com 21
- A Short Lesson in Simpler Prompts 2025
- AI-assisted coding for teams that can't get away with vibes 2025
- Artisanal shims for the bitter lesson age 2025
- Building the systems that build the software 2026
- Checking my model vibes against SWE-Bench Pro 2026
- Codex CLI vs Claude Code on autonomy 2026
- Engineering Maturity is all you need 2026
- Fight context rot with context observability 2025
- How I keep up with AI progress 2025
- How System Prompts Define Agent Behavior 2026
- How System Prompts Reveal Model Biases 2026
- Keeping up with ai
- Minimum Viable Benchmark 2025
- nilenso blog
- Our approach to evals with megasthenes 2026
- Sandboxing Megasthenes 2026
- The common sense unit of work 2025
- The quality of AI-assisted software depends on unit of work management 2025
- Trajectory shapes 2026
- Weird system prompt artefacts 2026
- What are popular AI coding benchmarks actually measuring? 2025
productcompass.pm 20
- Agent Loops for PMs: 20+ You Can Run This Week 2026
- Claude Code for PMs: The Beginner's Guide 2026
- Claude Code Pricing: Subscriptions vs API, Token Visibility, and the Models That Actually Work 2026
- Claude Code's Limits Are Generous. The Problem Is Your Setup. 2026
- Claude Dynamic Workflows for PMs: The Ultimate Guide 2026
- Claude Fable 5 Prompting: 5 Shifts From Measured Experiments 2026
- Claude Fable 5: The Ultimate Guide for PMs v3 (Paweł Huryn, 2026-06-11) 2026
- Claude Opus 5: The Best Opus Yet, Once You Heal It (Paweł Huryn, 2026-08-10) 2026
- How to Cut Your AI Bill From $200 to $20 a Month (Paweł Huryn, 2026-08-02) 2026
- How to Set Up Codex as a PM and Run It Next to Claude 2026
- I Don't Review the Code. I Review the Artifacts. 2026
- Product Engineering for PMs, Part 1: Build a SaaS App Without Coding 2026
- Product Engineering for PMs, Part 2: Build a SaaS App Without Coding 2026
- Stop Overpaying for Your AI Subscription: GPT-6.1 Sol, Sonnet 5.5, Opus 5.5 2026
- The Guide to Claude Code for PMs 2026
- The Intent Engineering Framework for AI Agents (Paweł Huryn, 2026-01-13) 2026
- The Ultimate Guide to Claude Opus 4.7 2026
- Three CLAUDE.md Blocks That Make Claude Get Smarter Every Session (Paweł Huryn, 2026-03-31) 2026
- What I Learned Building a Self-Improving Agentic System with Claude 2026
- Your .claude/ Folder Is a Production Agent 2026
steipete.me 17
- Claude Code Anonymous | Peter Steinberger 2025
- Claude Code is My Computer | Peter Steinberger 2025
- Command your Claude Code Army, Reloaded | Peter Steinberger 2025
- Commanding Your Claude Code Army | Peter Steinberger 2025
- Just One More Prompt | Peter Steinberger 2025
- Just Talk To It - the no-bs Way of Agentic Engineering | Peter Steinberger 2025
- llm.codes: Make Apple Docs AI-Readable | Peter Steinberger 2025
- MCP Best Practices | Peter Steinberger 2025
- My Current AI Dev Workflow | Peter Steinberger 2025
- OpenClaw, OpenAI and the future | Peter Steinberger 2026
- Posts
- Rss.xml
- Shipping at Inference-Speed | Peter Steinberger 2025
- Signature flicker
- Slot Machines for Programmers: How Peter Builds Apps 20x Faster with AI | Peter Steinberger 2025
- Stop Over-thinking AI Subscriptions | Peter Steinberger 2025
- The Future of Vibe Coding: Building with AI, Live and Unfiltered | Peter Steinberger 2025
huytieu.com 15
- A Release Video Skill From One Opus 5.5 Session · Huy Tieu 2026
- Agent Frameworks Are Converging on My Janky Markdown Setup · Huy Tieu 2026
- Agent Plugins: OpenAI, Microsoft, Amazon, Cursor and Vercel Agreed on One Plugin Format · Huy Tieu 2026
- AI Slop After the Em Dash · Huy Tieu 2026
- Anatomy of a Claude Code Conversation Transcript · Huy Tieu 2026
- Anthropic Dropped Opus 5. Here Is What You Should Update. · Huy Tieu 2026
- Designing for Personas Without the Theater · Huy Tieu 2024
- Fable 5.1, Opus 5 and Opus 5.5 on a Month of My Own Work · Huy Tieu 2026
- I Gave Claude Code My Entire QA Job · Huy Tieu 2026
- Scaling the Harness: Auditing My Personal Agent Against a Berkeley Paper · Huy Tieu 2026
- Start Simple, Scale With Intention · Huy Tieu 2025
- TermDeck: The Deck My Agents Work On · Huy Tieu 2026
- Total Internal Reflection · Huy Tieu 2026
- We Built Agile to Manage Slow Building. AI Just Made Building Fast. · Huy Tieu 2026
- Writing · Huy Tieu 2026
blog.sshh.io 13
- AI Can't Read Your Docs 2025
- AI-powered Software Engineering 2024
- Archive
- Building Multi-Agent Systems 2024
- Building Multi-Agent Systems (Part 2) 2025
- Building Multi-Agent Systems (Part 3) 2026
- Designing Software for Software Factories 2026
- Everything Wrong with MCP 2025
- How Cursor (AI IDE) Works 2025
- How I use AI in 2026 (Coding, Writing, Learning, Assistant-ing) 2026
- How I Use Every Claude Code Feature 2025
- I Asked 100 Agents to Hack Me 2026
- The Harness Is the Company 2026
newsletter.kentbeck.com 13
- Augmented Coding Technique: Copy From Simpler Language (Kent Beck, 2025-06-06) 2025
- Augmented Coding: Coached by the Genie 2025
- Baking a Model 2026
- Copy From a Simpler Language (Kent Beck, 2025-05-31) 2025
- Genie Fight 2025
- Genie Lessons from Genie Sessions: Prose as a Programming Language 2026
- Genie Lessons: Nobody Wants Agents (Kent Beck, 2026-04-23) 2026
- Genie Tarpit (Kent Beck, 2026-04-29) 2026
- Newsletter.kentbeck.com
- Passing Tests Bore Me (Kent Beck, 2026-04-22) 2026
- Starving Genies 2026
- Taming the Genie: "Like Kent Beck" 2026
- The Cost YAGNI Was Never About (Kent Beck, 2026-06-25) 2026
leehanchung.github.io 10
- Agentic AI Systems Explained with Super Mario 2026
- Claude Skills Deep Dive: How Agent Skills Work 2025
- Hidden Technical Debt of AI Systems: Agent Harness 2026
- Hidden Technical Debt of AI Systems: Agent Runtime 2026
- Poking Around the Claude Code Agent 2025
- Solve AI Engineering Slop with Machine Learning 2025
- Vibe Coding 101 for Software Engineers 2025
- Walkthrough: Setting Up Claude Code on Amazon Bedrock 2025
- Why AI Agents Are Workflows, Not Products 2025
- Why MCP Is Not a REST API for Agents 2025
scottspence.com 10
- Agentic engineering: a practical guide to reliable coding agents - Scott Spence 2026
- Coding agent harnesses with my-pi - Scott Spence 2026
- Give Coding Agents Your Project Docs With node:sqlite and FTS5 - Scott Spence 2026
- How I work with LLMs - Scott Spence 2026
- How to Make Claude Code Skills Activate Reliably - Scott Spence 2025
- I Built 21 MCP Tools. These Are the 2 I Still Use - Scott Spence 2026
- Measuring Claude Code Skill Activation With Sandboxed Evals - Scott Spence 2026
- omnirecall - Search Your Claude Code, Codex and Pi Sessions in One Place - Scott Spence 2026
- What Happened to My Coding Agent Guardrails After I Left - Scott Spence 2026
- What my prompts say about how I work with AI - Scott Spence 2026
lucumr.pocoo.org 8
blog.jonudell.net 5
grahamdumpleton.me 5
hadijaveed.me 5
- A Few Weekends with a Coding Agent Replaced My $30/Month Email App - Hadi Javeed's blog 2026
- Fable 5 Without the Token Anxiety - Hadi Javeed's blog 2026
- Plan Big, Execute Small: Taming My Fable 5 Bill Without Dumbing It Down - Hadi Javeed's blog 2026
- The Learning Loop: A Data Agent That Audits Its Own Answers - Hadi Javeed's blog 2026
- The LLM Router Era Starts with Fable 5 - Hadi Javeed's blog 2026
hamel.dev 5
aicoding.leaflet.pub 4
coderabbit.ai 4
dev.to 4
- Anthropic deleted 80% of Claude Code's system prompt. Here is what I changed in my setup. 2026
- Do your agent system prompts do anything? I measured 19 of mine 2026
- I Built a MITM Proxy to See What Claude Code Actually Sends to Anthropic 2026
- Sniffing Claude Code's API Calls: What Your IDE Is Really Sending 2026
every.to 4
ghuntley.com 4
lalitm.com 4
latent.space 4
eugeneyan.com 3
karpathy.bearblog.dev 3
mariozechner.at 3
note.com 3
claudecodecamp.com 2
codex.danielvaughan.com 2
explainx.ai 2
karpathy.ai 2
- HN Time Capsule 2015
- Jobs
marclove.com 2
matteo.vaccari.name 2
medium.com 2
metr.org 2
rlancemartin.github.io 2
sketch.dev 2
sonarsource.com 2
steve-yegge.medium.com 2
tidyfirst.substack.com 2
vals.ai 2
academy.dair.ai 1
- LLM Knowledge Bases 2026
adiasg.com 1
adr.github.io 1
agentconn.com 1
agenticloopsai.substack.com 1
ai.joaoqueiros.com 1
aibyaakash.com 1
- Claude Code 5. 2026
aihero.dev 1
aiquill.substack.com 1
alexmcfarland.substack.com 1
allenpike.com 1
ampcode.com 1
- Oracle 2025
antigma.ai 1
augmentcode.com 1
automationpanda.com 1
awesome-testing.com 1
besthub.dev 1
blog.cloudflare.com 1
blog.ctxinf.com 1
blog.promptlayer.com 1
blog.thepete.net 1
blog.trailofbits.com 1
botmonster.com 1
bughunt.productcompass.pm 1
- Bug Hunt Bench site 2026
c-sharpcorner.com 1
chadfowler.com 1
charity.wtf 1
charlesjones.dev 1
claude.ai 1
crawshaw.io 1
creatoreconomy.so 1
cursor-alternatives.com 1
d197for5662m48.cloudfront.net 1
designproject.io 1
dev.classmethod.jp 1
developersdigest.tech 1
donnamagi.com 1
dutchstartup.ai 1
elliotmilco.substack.com 1
embracethered.com 1
engineeredintelligence.substack.com 1
finance.biggo.com 1
flaviocopes.com 1
gertlabs.com 1
go.tabbit.ai 1
greptile.com 1
- Greptile Blog 2026
humanlayer.dev 1
huyenchip.com 1
- Blog 2025
huytieu.substack.com 1
indiehackers.com 1
interconnected.org 1
interconnects.ai 1
itmeetsot.eu 1
jamesshore.com 1
jduncan.io 1
john-rush.com 1
jxnl.co 1
karozieminski.substack.com 1
kentbeck.com 1
kirill-markin.com 1
kirshatrov.com 1
learn.chatgpt.com 1
lennysnewsletter.com 1
letta.com 1
linas.substack.com 1
llmjudge.com 1
lmarena.ai 1
lucadidomenico.studio 1
microsoft.com 1
millicosm.substack.com 1
morphllm.com 1
navigatingnoise.substack.com 1
nerdleveltech.com 1
newsletter.pragmaticengineer.com 1
oneusefulthing.org 1
prompt.anashel.com 1
prompthub.substack.com 1
qodo.ai 1
- Blog 2025
redreamality.com 1
ruben.substack.com 1
- Opus 5.5 2026
seangoedecke.com 1
sh-reya.com 1
sidsaladi.substack.com 1
substack.com 1
tbench.ai 1
- TERMINAL-BENCH 2026
techtimes.com 1
torqsoftware.com 1
utkarshkanwat.com 1
vintagedata.org 1
- Vintage Data 2026