Borrowing it
Nothing to install: this file belongs to RainFlashPoint/capital-agent-skills. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/RainFlashPoint/capital-agent-skills/main/CLAUDE.mdgit clone --depth 1 https://github.com/RainFlashPoint/capital-agent-skillsWrote 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.
[](https://agentmods.dev/instructions/rainflashpoint/capital-agent-skills/claude-md)<a href="https://agentmods.dev/instructions/rainflashpoint/capital-agent-skills/claude-md"><img src="https://agentmods.dev/badge/instructions/rainflashpoint/capital-agent-skills/claude-md.svg" alt="Measured on agentmods" height="20"></a>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 | $0.02118 | $0.02118 |
| Opus 5 | $0.01059 | $0.01059 |
| Sonnet 5 | $0.00424 | $0.00424 |
| Haiku 4.5 | $0.00212 | $0.00212 |
Grade A, and why
capital-agent-skills CLAUDE.md 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 7d 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.
How it starts
The opening of the file, as written. The whole thing — 50 lines — stays where its author put it; the contents beside it link to each section on GitHub.
capital-agent-skills — 维护契约(给在本仓工作的 coding agent 读)
本文件被 Claude Code 自动加载;
AGENTS.md(指向本文件)供 Codex 等读。 任何修改本项目的 agent,先读这里再动手。 用户向导见README.md;设计依据见docs/specs/。
这是什么
一套自洽、可移植、纯文件的 cap 研发流程技能族。结构、用法见 README.md。核心:Roles=职能视角(知识卡) / Skills=流程阶段;验证类(journey/model)是 cap-test 的 check 不是独立 skill。
三条不可破的铁律(改任何东西都要守)
- 不新增顶层 skill(除非是 cap 主线缺的真·生命周期阶段,如新增 release 那样的重大决策)。家族 =
cap-flowdriver(含 intake 需求树)+ 7 流程 skill(map / shape / plan / build / verify / review / release)+ harvest-experience。新增能力优先走:职能视角→references/roles/<r>.md;验证手法→cap-test/checks/<check>.md;语言→references/languages/<lang>.md;部署目标→cap-release/targets/<type>.md;流程纪律→references/*.md。真要加流程阶段=改 stage 枚举(driver + STATE 模板)+ driver 路由表 + 计数,慎重。 - 可移植:不硬依赖 Workflow / AskUserQuestion / subagent。交互用 text_mode(纯文本编号),并行用 Task-or-sequential 降级。必须在 Codex 下也能跑。
- 纯文件 + 单写者:知识/状态都是文件;
STATE.md单写者,并行产物各写各的。 - skill 约束"做什么/为什么/避哪些坑",不规定"怎么执行命令"。流程 skill 是约束与原则,不是命令手册——别写死带一堆 flag/转义/shell 怪癖的
find/grep/awk一行流(模型自己会写,写死只会脆、只会在某个 shell 崩)。把"坑"写成原则(如"统计大文件先排掉构建产物""别只扫根目录"),让模型自己选实现。例外:①git diff --name-only这类稳定且是契约输入的单命令可留;②references/languages/<lang>.md、cap-release/targets/<type>.md这类参考卡——它们的职责就是记录某语言/目标的具体 lint/test/build 命令,命令本身即交付物,保留。判据:这条命令是"流程怎么走"还是"某环境的具体工具调用"?前者→原则化,后者→留在参考卡。
怎么迭代(常见改动 → 要同步哪些地方)
| 改动 | 步骤(缺一即不一致) |
|---|---|
| 加一个角色卡 | ① 写 skills/cap-flow/references/roles/<role>.md(关注点/检查清单/好的样子/常见翻车/介入阶段)→ ② role-routing.md:§2 加触发规则 + §3 取值字典 + §5 自检 → ③ templates/STATE.md 角色取值字典 → ④ cap-understand Phase C 角色字典 → ⑤ 跑 scripts/validate-skills |
| 加一个 verify check | ① 写 skills/cap-test/checks/<check>.md → ② role-routing.md §4 字典 + §2 触发 → ③ cap-test 调度层引用 → ④ validate-skills |
| 加/改路由规则 | 只改 references/role-routing.md(它是改动→角色+check 的单一事实源);driver 只调用不内联 |
| 加一个流程阶段 | 慎重(要改 stage 枚举:cap-flow/SKILL.md + templates/STATE.md + 各 skill)。一般别加,优先用角色/check 扩展 |
| 改 shape/plan/build 等流程 | 改对应 skills/cap-<x>/SKILL.md;保持 §0 可移植前置不动 |
| 加/改 intake 需求树派生操作 | ① 在 scripts/intake.py 加子命令 + scripts/test_intake.py 加用例(TDD)→ ② cap-flow/references/intake.md 加操作章节 + 顶部 op 枚举 → ③ 若涉及特性生命周期(如 Retire 在 done 触发):driver §2/§4 + templates/STATE.md 同步 → ④ validate-skills。不新增 stage/skill(派生 op 挂在 cap-flow intake 下) |
| 加/改 meta 能力(改工具自身) | meta 能力(如 evolve 沉淀回流)不新增顶层 skill,用"角色卡 + playbook + driver 子命令"表达:references/roles/skill-maintainer.md(维护者视角)+ references/evolve-loop.md(回流 playbook)+ driver 的 /cap evolve 入口 + role-routing R10。小改走 /cap evolve(append-only),结构性大改对本仓跑完整 /cap |
| 改/扩 loop 编排 | loop(自治批量推进)同 evolve 范式:references/build-loop.md(playbook)+ driver /cap loop 入口,不新增 skill、不进 STATE 枚举、只编排既有阶段。往 build-loop.md append 一条原则(如 §E 蒸馏区)= 小改走 /cap evolve;改 loop 骨架/converge 契约 = 大改走完整 /cap |
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.
- 7d ago First seen · 50 lines · 2,118 tokens per session scan A 054a55314262
capital-agent-skills CLAUDE.md is an instructions file published in the GitHub repository RainFlashPoint/capital-agent-skills (2 stars, last pushed 3d ago), licensed MIT. It adds 2,118 tokens to every session, about $0.0106 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.
Other instructions, from other repositories
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.