i-have-adhd, a 50k-star Claude Code skill, makes AI agents lead with the answer
bigaiguy · x · 2026-09-26
A Claude Code skill called i-have-adhd has racked up 50,000 GitHub stars — not for improving code quality, but for fixing how the agent communicates. Most coding agents bury the actual fix under paragraphs of middleware context and dependency tangents, forcing you to dig for the answer every time.
The skill imposes 10 rules on the agent's output, including:
- Lead with the next action
- Number multi-step tasks
- End with exactly one concrete next step
- Give specific time estimates in minutes, not vague phrases
- No preamble, no recap, no closers
The README's before/after example says it all: the "after" version is one command, three numbered steps, and what to paste back if a test fails. The author argues this matters more than it looks, since response structure directly shapes human-agent collaboration speed.
More from coding & agent
- ChatGPT5.5 with a tool harness cracks Sokoban: 68 tool calls in 3 minutes — tak3sh8 · 2026-09-26
- Open-source APK Reverse Skill Hits 1.4k Stars, Lets Claude Code Analyze Android Apps — lxfater · 2026-09-26
- ThoughtDAG adds Jev-based context scoring: 391ms vs 24.8s for judging recalled excerpts — Lopsided_Scarcity979 · 2026-09-26
- Theo slams OpenRouter for using Jev: a non-reasoning classifier that can't gauge task complexity — intellectronica · 2026-09-26
- How Do You Optimize Your MCP Toolset So Agents Actually Use It Well? — onehundredemoji69 · 2026-09-26
- I Was Going to Ship 155 MCP Tools. The Token Math Said 10. — Difficult_Coffee_713 · 2026-09-26