ワンクリックでManusで任意のスキルを実行

taste

スター1

フォーク2

更新日2026年5月20日 21:02

Auto-bootstrap project taste.md + taste.vision via /deepresearch. NO 10-question wizard. Reads repo signals + researches SOTA at current time → synthesizes taste docs with zero questions (one fallback question only if repo has no signal).

インストール

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。

Manusで実行

ソース

waitdeadai

waitdeadai/claudemax

GitHub リポジトリを開く Creator のリポジトリを見る

ダウンロード

Manusで実行

/taste — auto-bootstrap project kernel

Replaces v1's /tastebootstrap (10 kernel questions). v2 auto-derives taste from repo evidence + current-time SOTA research.

Process

Read repo signals: README, package manifest, top-level structure, framework detection.
/deepresearch the project's SOTA at current time:
- " best practices "
- " SOTA architecture "
- WebFetch official docs of detected framework.
Synthesize (Opus) → taste.md (operating principles, code style, architecture invariants, deps posture) + taste.vision (north star, ICP, success criteria) with zero questions.
ONE clarifying question only if the repo has no signal (empty dir, no README, no package manifest). Exact question: "What are you building, in one sentence?"
Show the generated docs; user accepts or edits and re-runs.

Why auto-bootstrap

10 questions is friction. The model already has more signal from the repo + web than the user would type.
Anchored on current-time SOTA, not training-cutoff opinions.
Re-runnable cheaply — drift detection via /deepretaste is the same engine.

CLI

cmax taste init                          # auto-bootstrap; writes taste.md + taste.vision
cmax taste init --regenerate              # ignore existing; re-derive from scratch

What goes in taste.md vs taste.vision

taste.md — declarative rules. "We use X, not Y, because Z." Code style, architecture, deps.
taste.vision — narrative north star. One paragraph each: what we're building, who for, what "great" looks like.

How the harness uses taste

SessionStart hook reads taste.md and injects it into every workflow's system prompt.
/introspect cross-references plans against taste.md and flags deviations.
/audit and /review flag taste violations in existing code and proposed diffs.

このリポジトリの他の Skills

同じリポジトリ

ask

waitdeadai/claudemax

Ask. Achieve. The canonical user-facing entry to claudemax — describe your goal, the SOTA-2026 pipeline (deepresearch → multispec → parallel /goal → verify) handles the rest. Same engine as /cmax with a friendlier verb. CLI equivalent is `cmax ask "<goal>"`.

2026-06-091

cmax

waitdeadai/claudemax

The default FAT umbrella. Auto-runs deepresearch + multispec decomposition + parallel /goal + verify. Opus for planning/judgment/verification; Sonnet for sub-Spec execution (escalates per router). Multispec engine selects Mode A (SDK subagents) or Mode B (Claude Code Agent Teams) automatically based on work size. Your default daily-driver.

2026-06-091

opusolo

waitdeadai/claudemax

Max-effectiveness umbrella. Opus for EVERYTHING — planning, decomposition, sub-Spec /goal execution, verification. Use for novel domains, security/auth/payments, or when you want the highest ceiling regardless of cost.

2026-06-091

route

waitdeadai/claudemax

Show the model-routing decision for a task (Fable/Opus/Sonnet/Haiku) with reasoning. Use before launching a packet to sanity-check the tier and to override when you have stronger judgment than the heuristic.

2026-06-091

review

waitdeadai/claudemax

Reviews a proposed diff for correctness, security, style, taste alignment. Distinct from /audit (which reviews existing code, not diffs) and /verify (which checks SPEC completion).

2026-06-091

specqa

waitdeadai/claudemax

Spec quality gate. Every completion condition must have a mechanically-checkable verifyHint (file path, command, test name, behavior). Blocks /goal handoff if the spec wouldn't be verifiable.

2026-06-021

/taste — auto-bootstrap project kernel

Replaces v1's /tastebootstrap (10 kernel questions). v2 auto-derives taste from repo evidence + current-time SOTA research.

Process

Read repo signals: README, package manifest, top-level structure, framework detection.

/deepresearch the project's SOTA at current time:

" best practices "
" SOTA architecture "
WebFetch official docs of detected framework.

Synthesize (Opus) → taste.md (operating principles, code style, architecture invariants, deps posture) + taste.vision (north star, ICP, success criteria) with zero questions.

ONE clarifying question only if the repo has no signal (empty dir, no README, no package manifest). Exact question: "What are you building, in one sentence?"

Show the generated docs; user accepts or edits and re-runs.

Why auto-bootstrap

10 questions is friction. The model already has more signal from the repo + web than the user would type.

Anchored on current-time SOTA, not training-cutoff opinions.

Re-runnable cheaply — drift detection via /deepretaste is the same engine.

CLI

cmax taste init # auto-bootstrap; writes taste.md + taste.vision cmax taste init --regenerate # ignore existing; re-derive from scratch

What goes in taste.md vs taste.vision

taste.md — declarative rules. "We use X, not Y, because Z." Code style, architecture, deps.

taste.vision — narrative north star. One paragraph each: what we're building, who for, what "great" looks like.

How the harness uses taste

SessionStart hook reads taste.md and injects it into every workflow's system prompt.

/introspect cross-references plans against taste.md and flags deviations.

/audit and /review flag taste violations in existing code and proposed diffs.

name	taste
description	Auto-bootstrap project taste.md + taste.vision via /deepresearch. NO 10-question wizard. Reads repo signals + researches SOTA at current time → synthesizes taste docs with zero questions (one fallback question only if repo has no signal).