https://cuigh.com/tags/developer-tools/ Recent content in Developer Tools on AI Pulse Hugo en-US Mon, 13 Jul 2026 09:51:14 +0800 https://cuigh.com/posts/openai-gpt-5-6-prompt-guidance-2026/ Mon, 13 Jul 2026 09:51:14 +0800 https://cuigh.com/posts/openai-gpt-5-6-prompt-guidance-2026/ On 7/12, OpenAI quietly posted an official GPT-5.6 Sol prompt guidance on its developer site. The pithy data point: on OpenAI's internal coding-agent eval, lean system prompts lift scores 10–15%, cut tokens 41–66%, cut cost 33–67%. This piece isn't about 'the model is smarter so write shorter.' It's about a prompt-design philosophy reversal in the GPT-5.6 era — from 'enumerate rules to keep the model on rails' to 'describe the destination only, leave the path choice to the model.' https://cuigh.com/posts/agent-economy-three-layer-protocol-stack-2026/ Tue, 30 Jun 2026 09:29:00 +0800 https://cuigh.com/posts/agent-economy-three-layer-protocol-stack-2026/ Pulled from Product Hunt monthly, GitHub monthly, Hacker News weekly, and Follow Builders top tweets, as of June 30 the agent category has grown three independent tracks in the 'agent-talks-to-the-world' dimension: comms (Upstream), publishing (Publora), and settlement (Bluerails). Plus Panniantong/Agent-Reach gaining 25k stars in a month to crack the data wall, and @rauchg's June 28 'you don't need a LinkedIn' tweet (6.2k likes). This post argues the real inflection point of 2026 is not bigger models, it is that the protocols around agents are growing up the same way the protocols around the web did in 1995. https://cuigh.com/posts/agent-tools-five-layer-map-2026/ Sun, 28 Jun 2026 11:55:00 +0800 https://cuigh.com/posts/agent-tools-five-layer-map-2026/ Pulled from GitHub monthly leaderboard (star-history.com), Product Hunt monthly leaderboard, and HN weekly, as of June 28 the agent category has fractured into five independent tool layers: research / memory / orchestration / anti-slop / evaluation. Crossed with @rauchg on agent debuggability, @swyx on scaling without slop, and @nikunj on taste being earned in the arena, this piece argues the leverage has moved. It is no longer who writes the best prompt. It is who keeps their stack cleanly separated.