メニュー
Synthesize the settled conversation into a thin Plan on the work-tree leaf doc — no interview. Use when intent is pinned and you need a durable plan before building; helm's Plan phase reaches for this when the goal fits in one change.
デフォルトでは、最初にソースを確認する Prompt が選択されています。直接コマンドに切り替えるか、ローカルコピーをダウンロードすることもできます。
インストールを決める前に、SKILL.md と SkillsMP に表示されている付属ファイルをお読みください。
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
直接コマンドでは確認用 Prompt が省略されます。実行前にソースを確認してください。
npx skills add https://github.com/asteasolutions/ai-toolkit --skill to-specコマンドは1行のまま表示されます。コピー前に横へスクロールして全体を確認してください。
ローカルで確認しますか？SkillsMP が現在取得できるファイルをダウンロードできます。
Grounded research. Traceability-first research workflow for evidence, reliability, product, policy, technical, sentiment, complaint, review, public-opinion, and source-backed questions. Use when the user asks for grounded research, rigorous research, exact source traceability, text fragment links, citations, verification, fact-checking, adversarial review, counterevidence, confidence labels, source hierarchy, or a claim-to-source evidence map. Also use when another skill needs source-backed claims checked against primary sources, current product/tool information, community sentiment, or contested evidence.
Evidence discipline for factual claims. Receipts before claims, no guessing, unknown is an allowed answer. Use as the baseline standard when citing sources, files, or command output, and when the user asks you to re-check an answer you already gave because they cannot verify it themselves.
use when the user wants a simple worded reply, or expresses that your speech is too complex, jargon-heavy, hard to understand etc. Or when they say "keep it simple"
SKILL.md を表示中
| name | to-spec |
| description | Synthesize the settled conversation into a thin Plan on the work-tree leaf doc — no interview. Use when intent is pinned and you need a durable plan before building; helm's Plan phase reaches for this when the goal fits in one change. |
Turn settled conversation into a thin Plan. Do not interview — grilling already happened; capture what was decided.
Shapes match helm's work-tree artifacts (see helm/references/artifacts.md if present).
.scratch/<goal-slug>/ — this is the program counter.docs/agents/issue-tracker.md exists, follow its GitHub conventions for that mirror.Orient if needed. If the touched area isn't already understood, explore it. Prefer any domain vocabulary / decision docs the repo already has; discover them — don't invent a layout.
Sketch test seams. Prefer existing seams; pick the highest seam that still exercises the behaviour; minimise their number (ideal: one). Check seams with the developer before writing.
Write a thin Plan into the leaf doc:
.scratch/<goal-slug>/task.md## Plan section.Thin Plan contents only:
done criteriaNo brittle file paths or code. Exception: a decision-encoding prototype (schema, type, state machine) may be inlined when the shape is the decision.
Ask once about extras. After the thin Plan is written, ask whether to also include any of: user stories, out of scope, further notes (or other sections the developer names). Default is no. Add only what they pick.
GitHub mirror (opt-in). If publishing to GitHub, create/update the issue from the same Plan body. Local remains source of truth for resume.
# <feature title>
## Intent
<what + why>
## Plan
<what/why · test seams · implementation decisions · behaviour-level done criteria>
<!-- optional extras only if the developer asked for them -->