code-last-mile

code-last-mile is a skill for Claude Code from jtprogru/bear-skills. It costs 115 tokens per session (1,060 once invoked), scanned A, original, MIT.

A method for filling in the practical details that product plans often miss by describing one real user experience step by step. It focuses on what the user sees, does, waits for, and encounters when something goes wrong.

In plain words
What is it for?
It helps specify flows such as submitting an article and receiving connected notes, including interface states, waiting periods, errors, and other last-mile details.
Why use it?
It exposes usability gaps that remain hidden when a project is planned only as a list of technical components.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin.

Part of the bear-skills plugin — 42 skills, 3 commands, 11 agents, 1 hook shipped together

Good fit It helps specify flows such as submitting an article and receiving connected notes, including interface states, waiting periods, errors, and other last-mile details.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/jtprogru/bear-skills/code-last-mile
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 jtprogru/bear-skills --skill code-last-mile
Clone the repo
git clone --depth 1 https://github.com/jtprogru/bear-skills

Made for: Claude Code.

Or install bear-skills, the plugin that ships this one along with the rest of its 42 skills, 3 commands, 11 agents, 1 hook.

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 code-last-mile

README.md
[![agentmods](https://agentmods.dev/badge/skills/jtprogru/bear-skills/code-last-mile/github.svg)](https://agentmods.dev/skills/jtprogru/bear-skills/code-last-mile)
Your own site
<a href="https://agentmods.dev/skills/jtprogru/bear-skills/code-last-mile"><img src="https://agentmods.dev/badge/skills/jtprogru/bear-skills/code-last-mile/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 code-last-mile

Your own site · 80×15
<a href="https://agentmods.dev/skills/jtprogru/bear-skills/code-last-mile"><img src="https://agentmods.dev/badge/skills/jtprogru/bear-skills/code-last-mile.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 115 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,060 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 $0.00115 $0.01060
Opus 5 $0.00057 $0.00530
Sonnet 5 $0.00023 $0.00212
Haiku 4.5 $0.00012 $0.00106

Measured 8d ago against content hash e2a6e623379e, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

code-last-mile 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 8d 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.

domains/code/skills/code-last-mile/SKILL.md · 76 lines

How it starts

The opening of the file, as written. The whole thing — 76 lines — stays where its author put it; the contents beside it link to each section on GitHub.

code-last-mile — Последние 20%

Почему этот слой всегда выпадает

Когда задачу декомпозируют, декомпозируют существительное. «Сделать вики» распадается на ingestion, поиск, хранилище, слой LLM. Каждая часть выглядит необходимой, все они действительно нужны — и, будучи сделанными, дают систему, которой невозможно пользоваться.

Опыт использования не появляется в декомпозиции существительного. Значит, не становится задачей. Значит, не делается — и обнаруживается в конце как «работает, но что-то не то».

Лекарство: декомпозировать сцену, а не объект.

Шаг 1. Найди magic moment

Один конкретный момент, ради которого всё делается. Не «пользователь работает с системой», а сцена с деталями:

Человек вставляет ссылку на статью, уходит за кофе, возвращается — и видит, что статья уже разобрана на заметки, они связаны с тем, что он читал в прошлом месяце, и одна из связей его удивляет.

Если такую сцену не получается описать — это и есть главная находка. Продукта пока нет, есть набор компонентов.

Шаг 2. Разбери сцену по шагам

Пройди её как человек, а не как система. На каждом шаге отвечай:

  • что он видит
  • что он делает
  • что он ждёт
  • что он видит, если пошло не так

Последний вопрос вскрывает больше всего пропусков. В планах почти никогда нет ответа на «что показать, когда парсер не справился со статьёй».

Дополнительно по каждому шагу:

  • сколько он ждёт и понимает ли, что идёт работа
  • сколько действий нужно, чтобы дойти сюда с нуля
  • что он делает после — magic moment редко последний шаг

Шаг 3. Спецификация на уровне сантехники

То, что нашлось, описывается конкретно, а не пожеланиями. Не «улучшить обработку ошибок», а: «при таймауте парсера показать сообщение X, оставить ссылку в очереди, повторить через Y минут, при третьей неудаче написать в лог Z и показать пользователю кнопку W».

Уровень детализации такой, чтобы по описанию можно было реализовать, не додумывая. Формулировка, допускающая додумывание, будет додумана иначе, чем ты имел в виду.

Read the full file on GitHub · 76 lines

Files

What ships with it

1 file 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. 8d ago First seen · 76 lines · 115 tokens per session scan A e2a6e623379e

Subscribe to this mod's changes

code-last-mile is a skill published in the GitHub repository jtprogru/bear-skills (1 stars, last pushed 21d ago), licensed MIT. It adds 115 tokens to every session and 1,060 once invoked, about $0.0006 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.

Related

Other skills, from other repositories

sxo

Search Experience Optimization (SXO) — the bridge between SEO and UX/CRO. Audits the full journey from the SERP click to the on-page goal: SERP click-through factors (title/meta/rich results that win the click), then post-click experience signals that keep users and drive conversions — above-the-fold relevance and…

nowork-studio/notfair-plugin · 235 tokens

html-demo

A workflow for building a self-contained interactive HTML demo from a screenshot and a short requirement. The demo imitates the interface and includes simulated controls so its states can be explored.

ryanzhao1011/workframe · 85 tokens

frontend-design

Build web pages, components, landing pages, dashboards, or UI layouts. Use when asked to "build a page", "create a component", "design a landing page", "make a dashboard", or any frontend/UI implementation task.

runnithan/claudex-setup · 50 tokens

brand-and-identity

Use when establishing brand foundations before any design work, when two products need to feel related, when defining voice and tone for microcopy, or when logo usage and visual identity rules need documenting.

aneja5/forge-skills · 43 tokens

design-system

Use when establishing visual foundations for a new product, when defining design tokens, when building or auditing a component library, before any UI work begins on a fresh codebase, or when the existing UI has drifted into hex codes and magic numbers.

aneja5/forge-skills · 52 tokens

accessibility

Use when building UI that ships to users, when reviewing for WCAG AA compliance, when adding keyboard navigation, when ensuring screen reader support, when auditing color contrast, or when preparing a UI surface for an EU or government customer.

aneja5/forge-skills · 49 tokens