A verification coordinator for software changes. It selects and runs checks such as logic tests, user-journey tests, or AI-quality evaluations; TDD means testing during implementation, while this stage checks the finished change as a whole.

Use it after implementation to choose appropriate checks, run the required playbooks, assess user flows or model changes, and record the results.

It turns a claim that the change should work into documented evidence from a structured verification process.

cap-test — Skill for Claude Code by RainFlashPoint

Install

Getting it into your agent

It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.

Clone the repo

git clone --depth 1 https://github.com/RainFlashPoint/capital-agent-skills

agentmods

npx agentmods add skills/rainflashpoint/capital-agent-skills/cap-test

Made for: Claude Code.

Or install capital-agent-skills, the plugin that ships this one along with the rest of its 10 skills.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/skills/rainflashpoint/capital-agent-skills/cap-test.svg)](https://agentmods.dev/skills/rainflashpoint/capital-agent-skills/cap-test)

Your own site

<a href="https://agentmods.dev/skills/rainflashpoint/capital-agent-skills/cap-test"><img src="https://agentmods.dev/badge/skills/rainflashpoint/capital-agent-skills/cap-test.svg" alt="Measured on agentmods" height="20"></a>

Per session 313 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 5,730 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.00313	$0.05730
Opus 5	$0.00156	$0.02865
Sonnet 5	$0.00063	$0.01146
Haiku 4.5	$0.00031	$0.00573

Measured 3d ago against content hash d6b36988fa81, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

cap-test scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/cap-test/SKILL.md · 304 lines

How it starts

The opening of the file, as written. The whole thing — 304 lines — stays where its author put it; the contents beside it link to each section on GitHub.

cap-test — 验证中枢（调度层）

入口先执行 ../cap/SKILL.md 的“研发上下文门禁”：.cap 检查不可跳过，知识库注入失败可记录后离线继续。

全程契约：开始实质工作前读取并执行 ../cap-flow/references/progress-protocol.md 与 ../cap-flow/references/task-reconnaissance.md。先播报当前动作和下一步；新任务没有新鲜 .cap/task-context.md 时，先调查当前仓库代码，不能只依赖 PROFILE。

同时读取 ../cap-flow/references/harness-action-protocol.md。本阶段本地报告是诊断与 Provider 输入；当平台 Action API 可用且已有精确 Commit 时，必须创建独立 Test Action 并等待 Server 判定。Skills/STATE 不得自行证明质量 Gate PASS。

涉及测试账号、外部参数或外部状态写入时，遵守任务侦察协议的“先取证，再询问”：先复用仓库与知识库已有测试资产；授权与影响边界已经明确时直接推进，只有真实缺口或风险才设置人工门禁。

你是验证阶段的调度员。verify 是夹在 build(实现) 与 review(评审) 之间的独立阶段,职责是:把 "应该能用"变成"已验证能用"的硬证据。

定位铁律(与 build 区分)

build 内的 TDD 是边写边红绿的单元循环(写一个测试→红→实现→绿)。

verify 是成体系地证明它真的能用:跑全量套件 + 覆盖率门控、走用户旅程、评 AI 质量。两者不重复:build 关心"这个函数对不对",verify 关心"整个改动作为一个系统对用户 / 对质量站不站得住"。

调度层铁律 你不亲自实测。你做四件事:① 解析改动→定验证项;② 按序加载并执行每个验证项的 playbook;③ 汇总各验证项门控成本阶段总门控;④ 写报告 + 回写 STATE 交接。具体怎么跑测、怎么截图、怎么打分,全在 checks/<check>.md——你读它、照它做,不把它的内容重抄进本文件。引擎 = 一个能 Read/Edit/Bash/Grep 的模型(+ journey 验证项用 Playwright MCP)。

0. 可移植前置（每次入口先做）

共享 references 的位置:本文引用的 role-routing.md 物理上在编排器目录 cap-flow/references/ 下;而各验证项 playbook checks/<check>.md 就在本阶段目录(cap-test/checks/)下。解析路径分别指向 cap-flow/references/... 与本目录的 checks/...,别混。

本阶段要在 Claude 和 Codex 上都能跑。两条降级范式贯穿全程:

0.1 交互降级 —— 纯文本编号选项

凡需用户决策(选 scope、确认 eval 契约缺失时如何处理、确认 escalate),优先用纯文本编号列表:

本次 verify 要验的范围,选一个:
  1) feature   —— 只验本特性改动的旅程(默认)
  2) iteration —— 验本迭代累积的几处改动
  3) full-chain —— 全链路现状审计(也充当 cap-understand 的 baseline 健康检查)
回个编号即可。

宿主有结构化提问控件可用,但回退路径必须是上面这种编号文本。默认按编号文本写提示。

0.2 并行降级 —— 能并行就并行，不能就串行

当本轮要跑多个验证项(如 logic + journey + model)时,各产物互相独立、可并行:

有并行能力 → 可 fan-out,每个验证项写各自的报告文件(verify/<check>-...-report.md),最后汇总。
无并行能力 → 串行逐验证项执行同一份 playbook,逐个写报告。

无论并行还是串行,汇总写 STATE 由本调度层单点完成(单写者原则,见 §6)。

1. 入口条件（什么时候进 verify）

满足任一即可进入本阶段:

入口	条件	来源
主线推进	build 阶段门控通过(实现 green),STATE.stage 推进到 `verify`	cap-flow 路由
续接	STATE.stage 已是 `verify`(上次卡在某验证项)	跨会话 / 子 agent 接力
单项点名	用户 `/cap 测试 --check=<check> [--scope=<scope>]`	直接调用
现状审计	`--check=journey --scope=full-chain`(无需经 build)	cap-understand 的 baseline 健康检查

Read the full file on GitHub · 304 lines

Files

What ships with it

3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

3d ago Changed · +2 lines d6b36988fa81
7d ago First seen · 302 lines · 313 tokens per session scan A 250ccfcf7a90

Subscribe to this mod's changes

cap-test is a skill published in the GitHub repository RainFlashPoint/capital-agent-skills (2 stars, last pushed 3d ago), licensed MIT. It adds 313 tokens to every session and 5,730 once invoked, about $0.0016 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Testing

research-engineer

An uncompromising Academic Research Engineer. Operates with absolute scientific rigor, objective criticism, and zero flair. Focuses on theoretical correctness, formal verification, and optimal implementation across any required technology.

davila7/claude-code-templates · 43 tokens

tika-eval-compare

Compare extracts from two Tika builds over a corpus to detect regressions in content, encoding, exceptions, and embedded-document handling. Use for "compare before/after extracts", "eval this change against the corpus".

apache/tika · 50 tokens

neuron-evaluation-engineer

Create and run AI evaluations with datasets, assertions, and output drivers in Neuron AI. Use this skill whenever the user mentions evaluation, testing AI systems, creating evaluators, dataset-driven testing, assertion-based validation, or wants to measure AI system performance. Also trigger for tasks involving…

neuron-core/neuron-ai · 77 tokens

cap-test