lilmgenius/paperthin

re0-plan

Open a paperthin iteration's casebook before re0-loop's first turn — seeded with real content the moment the folder exists, never an empty directory.

ソースを見る

使い方

  1. インストールコマンドをコピーし、自分の環境へ導入する前にソースを確認します。
  2. 必要なツール、認証情報、対応環境を原文で確認します。
  3. 入力と成果物を指定し、出力を資料と照合します。
npx skills add https://github.com/lilmgenius/paperthin --skill re0-plan

開始前の確認

  • 導入にはモデル契約、API枠、ファイルへのアクセス許可は含まれません。
  • スター数や導入数は活動の指標であり、独立した品質・安全性試験ではありません。
リポジトリの原文

見出し、例、コード、表、リンク、参照画像を含む原文を表示しています。

Open .re0/iteration/<version>-<workname>/ and write into it in the same motion, before a single line of re0-loop's cycle runs.

Goal

A cycle without a written thesis, gate list, and step plan drifts to whatever framing comes first — and a folder opened with nothing in it tells a resumed session nothing at all. re0-plan closes both gaps at once: the folder is never empty from the moment it's created, regardless of which weight the cycle turns out to be. It assumes the full paperthin package installed and names its sibling skills directly. It pairs with re0-release, which retires the folder, and re0-memo, which extends whatever this skill seeded instead of restarting fresh. It is not a portable planning method — the shape it writes is this repo's own convention.

Workflow

  1. Classify the cycle's weight: a fix or hardening pass with no real design surface is lightweight; a cycle with a direction genuinely worth arguing through is full.
  2. Create .re0/iteration/<version>-<workname>/ and, in the same step, write its first file — the weight decides which:
  • lightweight → RETRO.local.md: one paragraph naming the task and why it's lightweight. re0-memo extends this same file at the end; it never starts a fresh one. Stop here.
  • full → DESIGN.local.md: thesis, scope, and quality gates — the same content re0-loop's FRAME turn expects, and enough on its own to survive a session break. When the change touches shippable surface, those gates always include an AGENTS.md/README reflection gate (see Rules). Confirm the read with readchk, size the tier with modelchk; reach for macrothink or another judgment skill only when the direction is genuinely contestable.
  1. (full only) Write WORKFLOW.local.md: the numbered steps this specific cycle runs through re0-loop's turn order — specific to this build, not a restatement of re0-loop's generic stages.
  2. (full only) Write EVIDENCE.local.md: the gates from DESIGN.local.md, listed as proof surface still to fill — the bar, not results.
  3. Keep reference material flat as REF-<topic>.local.md; promote to a refs/ subfolder only once it multiplies past a couple of files.
  4. Hand off to re0-loop. Do not invoke re0-work here — a from-scratch restart is re0-loop's own mid-cycle call, surfaced by nba when warranted, never a setup step.

Rules

  • The folder is never empty at creation. Every path, lightweight or full, writes real content in the same motion the folder is made, so a session break at any point still leaves something to resume from.
  • Not portable, by design. Names other paperthin skills directly and assumes the full package installed — the opposite of the self-contained default every other skill here follows. It exists only to pair with re0-release and re0-memo, not to travel alone.
  • No padding. Lightweight gets no DESIGN/WORKFLOW/EVIDENCE, and its RETRO.local.md seed stays one paragraph — a task note, not a WORKFLOW.local.md in disguise. A judgment skill or re0-work never gets forced in "just in case."
  • The folder's `<version>` is provisional. Apply the same minor/patch discriminator re0-release uses — kind not size: is the change, against the artifact's own prior spec, a fix (patch) or a new capability a user newly reaches for (minor)? — and keep it contingent on any unresolved fork the cycle opens. Never pre-commit a firm bump the cycle hasn't earned; a fork that could add a skill leaves the bump undecided until it resolves.
  • Docs stay aligned, like `re0-release` at ship time. When the cycle's change touches shippable surface, its quality gates always include an AGENTS.md/README reflection gate, assigning each reflection to its register per AGENTS.md's docs-role split: internal mechanism/rules/anatomy → AGENTS.md (LLM + human, injected as instructions), user-facing catalog/experience → README (human-only, never injected). Never dump agent-instruction mechanism into README, and never skip README for a user-facing change. This is the plan-time complement to re0-release's ship-time README step.
  • negatives-as-corpus — nothing this skill or the cycle it opens produces gets deleted; a cycle that goes wrong is retired by re0-release, not discarded.
  • Every EVIDENCE.local.md gate traces back to a line in DESIGN.local.md; don't invent proof surface DESIGN never asked for.

Verification

Before finishing:

  1. The folder has real content from the moment it exists — no empty directory, on either path.
  2. Full path: DESIGN.local.md states a thesis, scope, and gates; WORKFLOW.local.md's steps are specific to this cycle; every EVIDENCE.local.md gate traces to DESIGN.local.md.
  3. Lightweight path: RETRO.local.md is seeded with the task and reason, one paragraph.
  4. Reference material stayed flat unless it earned a refs/ subfolder.
  5. Report which path was taken and why.

タスクに関連する項目

同じリポジトリから

関連する Skills

すべての Skills
lilmgenius
コミュニティ

aim

When data lands with a thin ask — files, a repo, a paste, a chat scrollback — read it and propose what the user is trying to do with it, as a proposal to confirm rather than a question to answer. Use the moment material arrives without a fully-formed ask, or when the ask reads thinner than the data suggests.

1時間の導入数
1
GitHub Stars
1300
更新日
10月20日
lilmgenius
コミュニティ

detool

Make durable artifacts portable by replacing incidental stack, vendor, model, CLI, path, quota, or harness nouns with the mechanism they mean, while keeping provenance, runbooks, and tool-subject claims concrete.

1時間の導入数
1
GitHub Stars
1300
更新日
10月20日
lilmgenius
コミュニティ

feynman

Press a decision you just made until you can explain it to a skeptic in your own words, or surface the gap where you can't. The Feynman test aimed at a choice: understanding is cheapest to fake the moment after you decide, so a fresh critic presses for what you can't actually explain. Use right after picking an option, especially one an agent laid out without arguing for.

1時間の導入数
1
GitHub Stars
1300
更新日
10月20日
lilmgenius
コミュニティ

hate

Attack a plan, design, or argument like you want it to fail before you commit real effort — return the single load-bearing objection and the cheapest experiment that would prove it matters, not a checklist. User-invoked on purpose: a hate-it reflex always in the agent's reach would bias it toward demolition. Weighs many failure axes but owns the synthesis to one root.

1時間の導入数
1
GitHub Stars
1300
更新日
10月20日