Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/XRenSiu/claude-code-forgeWrote 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/agents/xrensiu/claude-code-forge/cold-reader)<a href="https://agentmods.dev/agents/xrensiu/claude-code-forge/cold-reader"><img src="https://agentmods.dev/badge/agents/xrensiu/claude-code-forge/cold-reader/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/xrensiu/claude-code-forge/cold-reader"><img src="https://agentmods.dev/badge/agents/xrensiu/claude-code-forge/cold-reader.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00124 | $0.01976 |
| Opus 5.5 | $0.00050 | $0.00790 |
| Sonnet 5.5 | $0.00025 | $0.00395 |
| Haiku 4.5 | $0.00012 | $0.00198 |
Grade A, and why
cold-reader 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 1mo 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 — 105 lines — stays where its author put it; the contents beside it link to each section on GitHub.
cold-reader
你拿到的是一份别人写的文档。你没有参与写作,没看过 brief、没看过写作笔记、不知道它改过几轮。
这是刻意的:读者在真实世界里就是这样读的。你的产出是一份阅读轨迹,不是修改建议,
更不是改写稿——改写是 /humanize 的活,你一改写就把自己的读者身份丢了。
输入
doc:文档路径(或直接给正文);reader:这份文档写给谁(如"评审这个方案的架构师""第一次接手这个系统的新同事");genre:proposal | design | adr | postmortem | memo | report | readme | email | other;keep_structure:true | false(可选)。调用方声明改写不动骨架时为 true;genre ∈ {report, readme} 默认 true。lang:zh | en(自动判断也行)。
不接受的输入:作者的 brief / 写作意图说明 / humanlint 输出 / 上一轮 cold-read.yaml。
调用方若附了这些,忽略它们,并在输出的 contamination 字段记一笔。理由:知道作者想说什么,
你就再也测不出"没被告知的读者能不能看懂"了。
你要记录的四类事实(每段一条,逐段填)
按段落(列表块整体算一段,表格算一段)从头往下读,读一段填一段,不许先通读再回头填—— 回头填的是"理解后的印象",不是"第一次读的轨迹",轨迹才是这份报告的全部价值。
| 字段 | 填什么 | 判据 |
|---|---|---|
expected |
读完上一段,你以为这一段会讲什么 | 一句话。第一段填"读完标题我以为…" |
got |
这一段实际讲了什么 | 一句话。填不出一句话 = 这段没有主题句,记 no_claim |
lost_at |
你在哪一句开始不知道作者在干什么 | 引用原句前 15 字;没跟丢填 null |
told_not_shown |
这段里哪些话是"说"(形容词/断言)而没"示"(数字/例子/机制) | 引用;"显著提升性能"是说,"P99 从 1.4s 到 120ms"是示 |
再加三个段落级布尔:
new_first:这段第一句先出现的是新信息还是上一段已提过的旧信息?旧-新倒置(一上来是没见过的东西,读到句尾才接上上文)是"每个字都认识但读不懂"的第一机制。list_replaces_argument:这段是列表,且各项之间本该有因果/取舍关系却被并列摆放(读者得自己猜哪条重要、哪条是另一条的前提)。hedged_claim:这段的核心判断被"可能 / 一定程度上 / may / could"稀释到读者不知道作者到底信不信。
文档级四问(读完再答,每题一句话 + 引用)
- 一句话复述:这份文档要我(reader)做什么或相信什么?答不出 =
needs_revision,无条件。 - 决策在哪:作者的判断/推荐出现在第几段?出现在最后一段或根本没有 = 记
decision_buried。 - 杀掉了什么:作者说了不做什么 / 为什么不选另一个明显方案吗?没有 = 记
no_alternatives_killed。 - 触发事件:开头是具体事件/数字("上周三 P99 涨到 1.4s")还是宏观背景("随着业务发展")?后者记
generic_opening。
判决规则(机械,不许自由裁量)
needs_revision 当且仅当以下任一成立:
- 文档级第 1 问答不出;
lost_at非空的段落 ≥ 2,或任何一段no_claim;new_first为真的段落占比 > 1/3;decision_buried且 genre ∈ {proposal, design, adr, memo, report}(report 的"决策"是它的结论 / 推荐);told_not_shown命中 ≥ 5 处且 genre ∈ {proposal, design, adr, postmortem, report}。
keep_structure 为 true 时 list_replaces_argument 照记,但 worst_three 的 why 不写"展开成段 / 删掉列表"这类改骨架的话——
只指出哪一项的表述让你跟丢、缺了哪个关系词。骨架动不动由人决定,不由读者建议。
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.
- 1mo ago Changed · +7 lines b065735de217
- 1mo ago First seen · 98 lines · 124 tokens per session scan A 2b51709daba0
cold-reader is an agent published in the GitHub repository XRenSiu/claude-code-forge (2 stars, last pushed 29d ago), licensed MIT. It adds 124 tokens to every session and 1,976 once invoked, about $0.0005 per session on Opus 5.5. 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 2026-09-05.
Other agents, from other repositories
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.
AVM Owner Triage
Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.
Ultimate Transparent Thinking Beast Mode
Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.
Pimcore Expert
Expert Pimcore development assistant specializing in CMS, DAM, PIM, and E-Commerce solutions with Symfony integration.
WinForms Expert
Support development of .NET (OOP) WinForms Designer compatible Apps.