load — for Claude Code community, for Claude Code, ide skills, handoff, resume, CLAUDE.md, ROADMAP.md, BACKLOG.md, git log --oneline -15, git status -sb

v1.0.0

このスキルについて

適した場面: Ideal for AI agents that need read state in this order. ローカライズされた概要: TANIK is a reference implementation of a multi-modal biometric kiosk # /load Run this at the start of a fresh chat after a previous session ran /handoff (and the user has typed /clear to wipe context). This AI agent skill supports Claude Code, Cursor, and Windsurf workflows.

機能

Read state in this order
CLAUDE.md — already auto-loaded; reaffirm the phase-gate + honesty conventions are top-of-mind
ROADMAP.md "Current status" — what phase, what was done, next action
BACKLOG.md — deferred items (don't act on them)
.claude/session-state.md if it exists — the previous session's task snapshot

# 主なトピック

Ayberkone Ayberkone
[0]
[0]
更新日: 4/25/2026

Skill Overview

Start with fit, limitations, and setup before diving into the repository.

適した場面: Ideal for AI agents that need read state in this order. ローカライズされた概要: TANIK is a reference implementation of a multi-modal biometric kiosk # /load Run this at the start of a fresh chat after a previous session ran /handoff (and the user has typed /clear to wipe context). This AI agent skill supports Claude Code, Cursor, and Windsurf workflows.

このスキルを使用する理由

推奨ポイント: load helps agents read state in this order. TANIK is a reference implementation of a multi-modal biometric kiosk # /load Run this at the start of a fresh chat after a previous session ran /handoff (and the user

おすすめ

適した場面: Ideal for AI agents that need read state in this order.

実現可能なユースケース for load

ユースケース: Applying Read state in this order
ユースケース: Applying CLAUDE.md — already auto-loaded; reaffirm the phase-gate + honesty conventions are top-of-mind
ユースケース: Applying ROADMAP.md "Current status" — what phase, what was done, next action

! セキュリティと制限

  • 制約事項: BACKLOG.md — deferred items (don't act on them)
  • 制約事項: READ-ONLY in this command. No Edit, no Write, no Bash that mutates state.
  • 制約事項: If ROADMAP.md "Current status" looks stale or wrong, say so explicitly — don't paper over it.

About The Source

The section below comes from the upstream repository. Use it as supporting material alongside the fit, use-case, and installation summary on this page.

Labs デモ

Browser Sandbox Environment

⚡️ Ready to unleash?

Experience this Agent in a zero-setup browser environment powered by WebContainers. No installation required.

Boot Container Sandbox

FAQ とインストール手順

These questions and steps mirror the structured data on this page for better search understanding.

? よくある質問

load とは何ですか?

適した場面: Ideal for AI agents that need read state in this order. ローカライズされた概要: TANIK is a reference implementation of a multi-modal biometric kiosk # /load Run this at the start of a fresh chat after a previous session ran /handoff (and the user has typed /clear to wipe context). This AI agent skill supports Claude Code, Cursor, and Windsurf workflows.

load はどうやって導入しますか?

次のコマンドを実行してください: npx killer-skills add Ayberkone/tanik。Cursor、Windsurf、VS Code、Claude Code など19以上のIDEで使えます。

load の主な用途は?

主な用途は次のとおりです: ユースケース: Applying Read state in this order, ユースケース: Applying CLAUDE.md — already auto-loaded; reaffirm the phase-gate + honesty conventions are top-of-mind, ユースケース: Applying ROADMAP.md "Current status" — what phase, what was done, next action。

load に対応するIDEは?

このスキルは Cursor, Windsurf, VS Code, Trae, Claude Code, OpenClaw, Aider, Codex, OpenCode, Goose, Cline, Roo Code, Kiro, Augment Code, Continue, GitHub Copilot, Sourcegraph Cody, and Amazon Q Developer に対応しています。統一された導入には Killer-Skills CLI を使えます。

load に制限はありますか?

制約事項: BACKLOG.md — deferred items (don't act on them). 制約事項: READ-ONLY in this command. No Edit, no Write, no Bash that mutates state.. 制約事項: If ROADMAP.md "Current status" looks stale or wrong, say so explicitly — don't paper over it..

このスキルの導入方法

  1. 1. ターミナルを開く

    プロジェクトディレクトリでターミナルまたはコマンドラインを開きます。

  2. 2. インストールコマンドを実行

    npx killer-skills add Ayberkone/tanik を実行してください。CLI がIDEまたはエージェントを自動検出し、スキルを設定します。

  3. 3. スキルを使い始める

    このスキルはすぐに有効になります。現在のプロジェクトで load をすぐ使えます。

! Source Notes

This page is still useful for installation and source reference. Before using it, compare the fit, limitations, and upstream repository notes above.

Upstream Repository Material

The section below comes from the upstream repository. Use it as supporting material alongside the fit, use-case, and installation summary on this page.

Upstream Source

load

Install load, an AI agent skill for AI agent workflows and automation. Explore features, use cases, limitations, and setup guidance.

SKILL.md
Readonly
Upstream Repository Material
The section below comes from the upstream repository. Use it as supporting material alongside the fit, use-case, and installation summary on this page.
Upstream Source

/load

Run this at the start of a fresh chat after a previous session ran /handoff (and the user has typed /clear to wipe context). Bootstraps fast and stops before doing any real work.

Naming note. Originally called /resume, but Claude Code has a built-in /resume that opens a session picker — the built-in always wins, so the project command never fires. Renamed to /load to avoid the collision.

Steps

1. Read state in this order

  1. CLAUDE.md — already auto-loaded; reaffirm the phase-gate + honesty conventions are top-of-mind
  2. ROADMAP.md "Current status" — what phase, what was done, next action
  3. BACKLOG.md — deferred items (don't act on them)
  4. .claude/session-state.md if it exists — the previous session's task snapshot

2. Reconstruct the working picture

Run in parallel:

  • git log --oneline -15 — recent commits, confirm against ROADMAP claims
  • git status -sb — should be clean (the previous /handoff refused otherwise)
  • git rev-parse HEAD — confirm current commit matches the one ROADMAP claims

If reality doesn't match what ROADMAP.md "Current status" or .claude/session-state.md claims (e.g., commits exist that the state file doesn't mention, or vice versa), stop and tell the user. Do not assume the state file or your reading is correct — surface the divergence so the user can resolve it.

3. Rebuild the TaskList

For every task listed in .claude/session-state.md (if present), recreate it via TaskCreate so the session-local task tracker matches what's pending. Preserve the [PHASE] prefix and the description/activeForm.

If .claude/session-state.md does not exist (fresh start without prior /handoff), skip this step — the user is starting clean.

4. Produce the load summary

Print, in five sentences:

  1. What TANIK is (one sentence — pulled from CLAUDE.md / README).
  2. Current phase + last commit SHA + subject.
  3. What was done last session (one or two bullet-style points if needed, in one sentence).
  4. What's pending and why it's pending (deploy decision? user action? in-flight bug?).
  5. The very next concrete action — name the task, name the file or files involved.

Then ask: "Continue with that, or different?"

5. Wait

Do not start any tool calls beyond the read-only ones above until the user replies. The point of /load is to load context, not to do work.

Rules

  • READ-ONLY in this command. No Edit, no Write, no Bash that mutates state.
  • If ROADMAP.md "Current status" looks stale or wrong, say so explicitly — don't paper over it.
  • If the user asks "what was the last thing you did?", trust git history over the state file (history is canonical, state file is a hint).
  • The TaskList rebuild is best-effort. If a task description in .claude/session-state.md doesn't make sense in the current context, ask before recreating it verbatim.

関連スキル

Looking for an alternative to load or another community skill for your workflow? Explore these related open-source skills.

すべて表示

openclaw-release-maintainer

Logo of openclaw
openclaw

ローカライズされた概要: 🦞 # OpenClaw Release Maintainer Use this skill for release and publish-time workflow. It covers ai, assistant, crustacean workflows. This AI agent skill supports Claude Code, Cursor, and Windsurf workflows.

333.8k
0
AI

widget-generator

Logo of f
f

ローカライズされた概要: Generate customizable widget plugins for the prompts.chat feed system # Widget Generator Skill This skill guides creation of widget plugins for prompts.chat . It covers ai, artificial-intelligence, awesome-list workflows. This AI agent skill supports Claude Code, Cursor, and Windsurf

149.6k
0
AI

flags

Logo of vercel
vercel

ローカライズされた概要: The React Framework # Feature Flags Use this skill when adding or changing framework feature flags in Next.js internals. It covers blog, browser, compiler workflows. This AI agent skill supports Claude Code, Cursor, and Windsurf workflows.

138.4k
0
ブラウザ

pr-review

Logo of pytorch
pytorch

ローカライズされた概要: Usage Modes No Argument If the user invokes /pr-review with no arguments, do not perform a review . It covers autograd, deep-learning, gpu workflows. This AI agent skill supports Claude Code, Cursor, and Windsurf workflows.

98.6k
0
開発者