The custom instructions didn't matter much. Claude followed them well: as you can see here, one conducted negotiations entirely in the persona of an exasperated, down-and-out cowboy. But "hardball Claudes" didn't generally fare better than "courteous Claudes."
PROMPT ENGINEERING
-
Anthropic’s Claude Code setup plugin simplifies AI automations
By
–
Claude Code es un lío. Hasta que instalas esto.
— Nico (@nicos_ai) 24 avril 2026
Hay un plugin oficial de Anthropic llamado claude-code-setup.
Te dice qué automatizaciones puedes montar (hooks, skills, MCP servers, subagentes…) y cómo configurarlas paso a paso.
Básicamente analiza tu proyecto y te recomienda… pic.twitter.com/ohQnpFSIE1Claude Code is a mess. Until you install this. There's an official Anthropic plugin called claude-code-setup. It tells you what automations you can set up (hooks, skills, MCP servers, subagents…) and how to configure them step by step. Basically, it analyzes your project and
-
Claude Distorts Topics to Match User’s Assumed Interests
By
–
Me: Can you tell me about [thing]? Claude: ah, yes. You’re probably asking because of your interest in [other thing], and honestly, those two things have more in common than you even realize. [Bastardized version of [thing] that contorts it to fit [other thing].]
-
Controlling AI output via prompts and verification for longer thinking
By
–
IME you still get a lot of control over this in the prompt itself. Even with the highest Pro settings, it thinks longer if you say to keep trying until some verification you propose passes etc.
-
Claude Dispatch with Computer Use for Codex Integration
By
–
anyone else using claude dispatch w/ computer use to use codex?
-
Claude Code Structures Output but Misses Contextual Business Insights
By
–
The analyzing is still kind of the hard part. Claude Code can structure the output but it can't tell you which insight actually matters for your specific situation.
-
Seven Core Design Decisions for Production Agent Harnesses
By
–
Design principles for building an Agent harness. Most agent builders get three of the seven core design decisions exactly backwards. Every production agent harness is the result of seven architectural bets. Agent count, reasoning strategy, context strategy, verification,
-
Queue prompts with mid-thinking correction steering
By
–
Just send a prompt and then it queues, then you can also send a steer prompt to correct it mid thinking, there's a button for that
-
Future LLMs Will Make Prompting as Simple as Asking
By
–
When the LLMs and agents have context, memory, and, eventually, continual learning, prompting will be as simple as asking for what you want. Or sometimes not even having to ask (predictive agents). https://t.co/STBCscbofG
— Paul Roetzer (@paulroetzer) 24 avril 2026When the LLMs and agents have context, memory, and, eventually, continual learning, prompting will be as simple as asking for what you want. Or sometimes not even having to ask (predictive agents).
-
Testing AI Skills and Plugins on Codex Platform
By
–
curious if you’ve tried your skills/ plugins on codex and how the experience has been!