ayghri/i-have-adhd

A skill to stop your coding agent from burying the answer. ADHD-friendly output.

View on GitHub ↗Jump to charts ↓Open shareable report

Summary Information

Updated 1 hour ago
Added to GitGenius on August 2nd, 2026
Created on May 13th, 2026
Open Issues & Pull Requests: 70 (+0)
GitHub issues: Enabled
Number of forks: 2,875
Total Stargazers: 49,893 (+27)
Total Subscribers: 154 (-1)

Repository Insights (GitGenius)

Median issue/PR response: 3.1 hours
Mean response time: 2.4 days
90th percentile: 8.8 days
Tracked items: 55

How this project is maintained

Roughly one issue in five opened in the past year never receives a reply. 96% of open issues come from outside the core team, so the backlog reflects real-world use rather than internal planning. 64% of issues opened in the past year have been closed, leaving a working backlog. Three people close 93% of everything that gets resolved.

Charts & Analytics

Fetching additional details & charts...

Issue Activity (beta)

Open issues: 25
New in 7 days: 5
Closed in 7 days: 2
Avg open age: 8 days
Stale 30+ days: 4
Stale 90+ days: 0

Recent activity

Opened in 7 days: 5
Closed in 7 days: 1
Comments in 7 days: 4
Events in 7 days: 6

Top labels

  • Target:Rules (2)
  • AI Agora (1)
  • Author:AI (1)
  • enhancement (1)
  • issue (1)
  • question (1)

Most active issues this week

Sign in to see which issues are moving.

Detailed Description

The i-have-adhd repository is a Python-based skill designed for Claude Code and other coding agents that reformats their output to be more direct and action-oriented. The core purpose is to prevent coding assistants from burying actionable answers beneath lengthy explanations, preamble, and conversational filler. The skill enforces a structured output format that prioritizes immediate next steps, numbered instructions, and concrete guidance without pleasantries or tangential discussion.

The repository implements ten specific rules that govern how a coding agent should respond when the skill is invoked. These rules mandate leading with the next action, numbering multi-step tasks, ending with one concrete next step, suppressing tangents, restating state each turn, providing specific time estimates in minutes rather than vague durations, making wins visible, handling errors matter-of-factly, capping lists at five items maximum, and eliminating preamble, recap, and closing pleasantries. The before-and-after example in the README demonstrates the transformation: instead of a conversational explanation about auth flows that mentions multiple files and suggests various approaches, the skill produces a direct command to run a package update, numbered steps to edit a specific file with line numbers, a test command to run, and a single clear next step if tests fail.

Installation varies by coding agent platform. For Claude Code users, the skill can be invoked with the command /i-have-adhd without requiring a local clone, as Claude Code automatically fetches and maintains the repository. Users can also configure the skill to activate on every session by creating a file at ~/.claude/.i-have-adhd-always. For Codex users, the skill is invoked with $i-have-adhd and can be applied both explicitly and implicitly when the system detects tasks that would benefit from the output style. Additional installation instructions for other coding agents are documented in INSTALL.md.

The repository is designed to be customizable. Users can fork the repository, edit the rules defined in skills/i-have-adhd/SKILL.md, and swap in their modified version by restarting Claude Code and re-invoking the skill. The full text of the ten rules is maintained in SKILL.md for reference and modification.

The skill's design is loosely based on The Adult ADHD Tool Kit by J. Russell Ramsay and Anthony L. Rostain, though it has been adapted specifically for how large language models should structure responses rather than how humans should organize their daily tasks. The repository is licensed under MIT and includes README translations in Simplified Chinese, Japanese, Korean, Vietnamese, and Brazilian Portuguese, indicating an international user base. The project encourages users to star the repository if the skill helps them avoid scrolling past unnecessary conversational preamble in coding assistant responses.