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.
npx skills add 0fuk/OpenACCP --skill human-explaingit clone --depth 1 https://github.com/0fuk/OpenACCPWrote 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/skills/0fuk/openaccp/human-explain)<a href="https://agentmods.dev/skills/0fuk/openaccp/human-explain"><img src="https://agentmods.dev/badge/skills/0fuk/openaccp/human-explain/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/skills/0fuk/openaccp/human-explain"><img src="https://agentmods.dev/badge/skills/0fuk/openaccp/human-explain.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 | $0.00079 | $0.00807 |
| Opus 5.5 | $0.00032 | $0.00323 |
| Sonnet 5.5 | $0.00016 | $0.00161 |
| Haiku 4.5 | $0.00008 | $0.00081 |
Grade A, and why
human-explain 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 2d 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.
What it actually says
Human Explain / 说人话
Purpose
把当前对话里的技术状态、工程术语、架构概念翻译成人能直接听懂的自然语言。本 skill 只管怎么说;汇报该包含什么、什么算证据,归 /formal-report。触发:/human-explain 或 /说人话。
Rules
- 自然中文短句,像同事面对面聊天。
- 结论先行:先说「所以怎样」,再补细节。
- 提到任何 ID、术语、文件名,先用一句话说它对产品/交付的意义,再给 ID;数字后面跟一句它对用户 / 业务 / 交付意味着什么。
- 状态不明确就说清哪里不明确,不编。
- 保留必要英文术语(PR、CI、merge、文件名、状态值),其余全中文。
- 业务语言铁律(owner 2026-09-06 打回后加):给 owner 的说明里一个代号都不许出现——工单号 / 任务编号、迁移号、commit 哈希、门次或阶段符号(①②之类)、内部代称(版本代号、基线代号、补丁代号)、分支或工作树名、角色或会话代号,一律翻成「它是什么、对用户 / 业务 / 交付意味着什么」的一句话:说「记录每个账号今天能看什么能做什么的那张基线表」不说基线代号,说「负责某某重构那条线」不说会话代号,说「过了第一道验收关(自测加独立复核)」不说门次符号。必须给路径或编号时(owner 要点开看、要转发),放在段末括号或末尾「关键路径」一节,正文里不带。落笔前自检:通读一遍,凡是一个不在项目里的人读不懂的词,当场替换;替换不了就删。实锤(2026-09-06):一整份汇报全是编号、哈希、阶段符号,owner 原话「说了等于白说,根本就不知道你做了什么」。
Forbidden patterns
- 「不是 X,而是 Y」「这不仅是 X,更是 Y」
- 「本质上」「从某种意义上说」「这意味着」
- 「鲁棒」「闭环」「抓手」「赋能」「沉淀」「打通」「对齐」「落地」「链路」
- 「让我们来看看」「接下来我将」「首先需要指出的是」
- 不用比喻、类比、修辞替代直接说明
Output shape
1-2 句结论 → 具体在干什么(每条一句)→ 为什么重要 → 下一步/需要谁做什么决定。不需要 header、分隔线、表格,除非内容确实需要对比。
Boundaries
解释模式:不写文件、不起 agent、不改状态。解释完再执行用户要的操作。信息不够就说「我需要看 X 才能确认」,不猜。
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.
- 2d ago First seen · 31 lines · 79 tokens per session scan A d3c02f2f7bb9
human-explain is a skill published in the GitHub repository 0fuk/OpenACCP (2 stars, last pushed 2d ago), licensed MIT. It adds 79 tokens to every session and 807 once invoked, about $0.0003 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 2026-10-14.
Other skills, from other repositories
sema-seed
Seed the sema vocabulary by exploring random words. Pull a word from the system dictionary, think about it through the sema lens without forcing it into a pattern, then mint only if a genuinely useful and distinct pattern emerges. Use when the user asks to "seed", "expand", "grow", or "explore new patterns for" the…
handoff-mesh
Built-in communication layer for multi-agent setups. Full-mesh handoff folders (any agent can hand off to any other), an hourly watcher that surfaces open handoffs, weekly archive cleanup, and a completion-reporting loop so every handoff is handled AND verified. The durable 'brain' counterpart to real-time channels…
oma-explanation
Create an offline HTML explanation of a code change (diff, PR, branch) or of a topic, system, or question. Use when a visual walkthrough document is requested.
moai-workflow-jit-docs
Enhanced Just-In-Time document loading system that discovers, loads, and caches relevant documentation based on user intent and project context. Use when users need specific documentation on demand.
agent-teams
Coordinates Claude agent teams via filesystem protocol. Use when orchestrating parallel agents with task dependencies. Do not use for single-agent tasks.
aiwg-guide
Contextual AIWG help — explains current version features, answers how-to questions, routes live queries to the steward.