一键在 Manus 中运行任何 Skill

开始使用

hima-warden

星标1

分支0

更新时间2026年7月2日 18:59

Prove a change works end-to-end by driving the real flow — the dynamic counterpart to hima-verdict's static review.

安装

用 Codex 或 Claude 帮你安装复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。

在 Manus 中运行

来源

momomuchu

momomuchu/hima

打开 GitHub 仓库查看创作者相关仓库

下载

在 Manus 中运行

Purpose

Exercise the actual runtime surface a change touches instead of trusting tests, a diff read, or a code review alone. hima-warden is the dynamic proof step: it drives the affected flow (UI, API, CLI) and observes real behavior before a completion claim is trusted.

When to use

Force at M+ for any change with a runtime surface (UI flow, API endpoint, CLI command).
Skip explicitly for docs-only or test-only diffs — there is nothing to drive.
Pairs with hima-verdict: verdict reviews statically, warden proves dynamically; neither substitutes for the other.

Steps

Identify the runtime surface the change actually touches (endpoint, screen, command).
Drive it for real — not a mock, not a dry-run: real request, real render, real invocation.
Walk the happy path first, then edge cases (empty/error states, boundary inputs).
Compare observed behavior against the charter's acceptance criteria, not against the implementer's own description of what it should do.
Record the evidence (command output, screenshot, transcript) that proves the flow ran.

Done when

The affected flow has been driven for real with recorded evidence, and the observed behavior matches (or explicitly deviates from, with a named gap) the charter's acceptance criteria.

同仓库更多 Skills

同仓库

hima-admit

momomuchu/hima

Bootstrap a new project or catch up an agent on an existing one in minutes by reading vision/goals/state and scaffolding the minimum brain files.

2026-07-021

hima-blueprint

momomuchu/hima

Produce the structural plan — component boundaries, contracts, data flow — plus a lightweight decision record, from a sealed charter.

2026-07-021

hima-charter

momomuchu/hima

Formalize intent into a spec artifact (Gherkin scenarios, acceptance criteria, criticality tags) before any non-md file may change.

2026-07-021

hima-covenant

momomuchu/hima

Declare and audit the per-project disciplines (cadence, branching, testing, architecture, methodology) that the pluggable StageSkillMap resolves against.

2026-07-021

hima-forge

momomuchu/hima

Implement the minimum code to satisfy the charter's RED tests, one logical change per commit.

2026-07-021

hima-lookout

momomuchu/hima

Cited, multi-source research pass for H/C decisions or true blockers — converts findings into requirements, risks, and rejected alternatives.

2026-07-021

name	hima-warden
description	Prove a change works end-to-end by driving the real flow — the dynamic counterpart to hima-verdict's static review.
stage	meta
source	base
mode	force

Purpose

When to use

Force at M+ for any change with a runtime surface (UI flow, API endpoint, CLI command).
Skip explicitly for docs-only or test-only diffs — there is nothing to drive.
Pairs with hima-verdict: verdict reviews statically, warden proves dynamically; neither substitutes for the other.

Steps

Identify the runtime surface the change actually touches (endpoint, screen, command).
Drive it for real — not a mock, not a dry-run: real request, real render, real invocation.
Walk the happy path first, then edge cases (empty/error states, boundary inputs).
Compare observed behavior against the charter's acceptance criteria, not against the implementer's own description of what it should do.
Record the evidence (command output, screenshot, transcript) that proves the flow ran.

Done when

The affected flow has been driven for real with recorded evidence, and the observed behavior matches (or explicitly deviates from, with a named gap) the charter's acceptance criteria.