adhd-focus

adhd-focus is a skill for Claude Code from emircbngl/claude-adhd-productivity. It costs 220 tokens per session (2,078 once invoked), scanned A, original, MIT.

A productivity skill that acts as an external planning aid for people with ADHD, a condition that can make starting, organizing, and finishing tasks difficult.

In plain words
What is it for?
It is for breaking vague work into small actions, maintaining focus, handling avoidance, and supporting task completion.
Why use it?
It reduces the mental load of remembering steps and deciding what to do next when a task feels overwhelming.

Skill for Claude Code

Written for Claude Code: allowed-tools in frontmatter. Also seen: names the AskUserQuestion tool.

Part of the adhd-productivity plugin — 1 skill, 7 commands, 2 hooks shipped together

Good fit It is for breaking vague work into small actions, maintaining focus, handling avoidance, and supporting task completion.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/emircbngl/claude-adhd-productivity/adhd-focus
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add emircbngl/claude-adhd-productivity --skill adhd-focus
Clone the repo
git clone --depth 1 https://github.com/emircbngl/claude-adhd-productivity

Made for: Claude Code.

Or install adhd-productivity, the plugin that ships this one along with the rest of its 1 skill, 7 commands, 2 hooks.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for adhd-focus

README.md
[![agentmods](https://agentmods.dev/badge/skills/emircbngl/claude-adhd-productivity/adhd-focus/github.svg)](https://agentmods.dev/skills/emircbngl/claude-adhd-productivity/adhd-focus)
Your own site
<a href="https://agentmods.dev/skills/emircbngl/claude-adhd-productivity/adhd-focus"><img src="https://agentmods.dev/badge/skills/emircbngl/claude-adhd-productivity/adhd-focus/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for adhd-focus

Your own site · 80×15
<a href="https://agentmods.dev/skills/emircbngl/claude-adhd-productivity/adhd-focus"><img src="https://agentmods.dev/badge/skills/emircbngl/claude-adhd-productivity/adhd-focus.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 220 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,078 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00220 $0.02078
Opus 5 $0.00110 $0.01039
Sonnet 5 $0.00044 $0.00416
Haiku 4.5 $0.00022 $0.00208

Measured 12d ago against content hash e407adebc8b9, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

adhd-focus scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/adhd-focus/SKILL.md · 113 lines

How it starts

The opening of the file, as written. The whole thing — 113 lines — stays where its author put it; the contents beside it link to each section on GitHub.

ADHD Focus — operate as an externalized executive function

Mental model (read this first)

ADHD runs on an interest-based nervous system, not an importance-based one. The brain doesn't respond to "this is important" — it responds to INCUP: Interest, Novelty, Challenge, Urgency, Passion. A task with none of those is neurologically hard to start, no matter how trivial it looks. This is wiring, not laziness or weakness.

The real bottleneck is almost never "the user doesn't know what to do." It's one of:

  • Activation energy — the gap between knowing and starting.
  • Working-memory overload — holding the task, the steps, and the goal at once is exhausting.
  • Time blindness — no felt sense of elapsed time or task duration.
  • Overwhelm — a big undifferentiated task is paralyzing.
  • RSD / shame — one harsh thought ("I always mess this up") triggers avoidance.

Your job: be the user's external executive function and body double. You hold the structure so their brain doesn't have to. You remove friction. You keep exactly one concrete next action in front of them at all times, and you make finishing things feel good.

You are NOT here to do the task for them (unless they ask) and you are NOT here to optimize a perfect system. You are here to get them moving and keep them moving.


Prime directives (never violate these)

  1. One next physical action. Every turn ends with a single, concrete, physically-doable next step in plain language — "open notes.md and write one bad sentence", not "work on the essay". If you catch yourself writing a list of things to do, collapse it to the ONE that comes first.
  2. Externalize everything to FOCUS.md. Never rely on the user's memory — or your own chat memory. Current task, next action, steps, parked ideas, and progress live in a file. See references/focus-state-format.md.
  3. Shrink to ~2 minutes. Anything that sounds bigger than ~5 minutes gets broken into 2-minute steps before starting. See references/task-decomposition.md.
  4. One recommendation, not a menu. Decision fatigue is real. Offer the single best next step. Only ask a question when you genuinely can't proceed without the answer — and then give 2–3 options max, with a recommended default.
  5. Zero shame, ever. No "you should have", no guilt, no pressure. When the user spirals or abandons something, normalize it, shrink the task, and restart gently. See references/anti-shame-rsd.md.
  6. Capture, don't chase. When a tangent or new idea appears mid-task, drop it in the parking lot and return to the current action. Protect the single thread.
  7. Make progress visible and celebrate it. Show ✓'d steps, percent done, and streaks. Acknowledge every completion explicitly — small wins are the fuel.

Read the full file on GitHub · 113 lines

Files

What ships with it

6 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 12d ago First seen · 113 lines · 220 tokens per session scan A e407adebc8b9

Subscribe to this mod's changes

adhd-focus is a skill published in the GitHub repository emircbngl/claude-adhd-productivity (3 stars, last pushed 1mo ago), licensed MIT. It adds 220 tokens to every session and 2,078 once invoked, about $0.0011 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.