#160160th of 193 in Skills, plugins and prompts
adhd
Skill that runs isolated parallel reasoning frames plus a critic pass for ideation
Documentation ↗
(opens in a new tab)Website ↗
(opens in a new tab)Repository on GitHub ↗
(opens in a new tab)
- Stars
- 4.4k
- License
- MIT
- Last commit
- Oct 2026
- Last release
- May 2026
- Language
- TypeScript
208 on 26 May 20264,444 on 20 Oct 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
ADHD is an agent skill, npm CLI and TypeScript library that spawns N isolated LLM calls, each under a distorted cognitive frame with no shared context. A separate critic call then scores ideas on novelty, viability and fit, flags traps, clusters them and deepens the top-K. It installs with `npx skills add` into Claude Code, Cursor, Codex, Cline, Gemini CLI and others.
Who it is for: Developers using coding agents for design decisions, naming and fuzzy debugging
Strengths
- Generator and critic are separate LLM calls, so the split is mechanical, not prompt-based
- Installs as a skill into Claude Code, Cursor, Codex, Cline, Gemini CLI and about 50 other agents
- Also ships as an npm CLI (adhd-agent) and library returning shortlist, traps and clusters
- Eval transcripts and methodology are in the repo (bench/results.json, documentation/evals.md)
Weaknesses
- Benchmarks are author-run: 6 problems scored by an LLM judge
- One external review measured about 2.3x time and 1.9x output versus single-shot
- Aimed at open-ended ideation; little use for routine coding tasks
- README does not say which model providers the CLI and library support
What it needs
- no GPU
- Needs Node.js >=18, LLM agent or API access
- Models: unknown
Also in Skills, plugins and prompts
See all 193| Rank | Project | Score |
|---|---|---|
| 1 | cavemanCuts token use in coding agents with terse replies and compressed inputs | 95 out of 100 |
| 2 | impeccableDesign skill with 24 commands and a detector for AI-generated frontends | 95 out of 100 |
| 3 | skillsAgent skills for planning, TDD, debugging and code review | 94 out of 100 |
| #4 | eccAgents, skills, hooks and rules bundle for Claude Code, Codex and other harnesses | 94 out of 100 |
| #5 | oh-my-claudecodeMulti-agent orchestration plugin and CLI for Claude Code | 93 out of 100 |