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 orch-plangit 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/orch-plan)<a href="https://agentmods.dev/skills/0fuk/openaccp/orch-plan"><img src="https://agentmods.dev/badge/skills/0fuk/openaccp/orch-plan/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/orch-plan"><img src="https://agentmods.dev/badge/skills/0fuk/openaccp/orch-plan.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 | $0.00144 | $0.05687 |
| Opus 5.5 | $0.00058 | $0.02275 |
| Sonnet 5.5 | $0.00029 | $0.01137 |
| Haiku 4.5 | $0.00014 | $0.00569 |
Grade A, and why
orch-plan 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.
How it starts
The opening of the file, as written. The whole thing — 213 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/orch-plan — 编排计划(方向 + 工程双层)
存在理由:20260901,owner 打回一版编排计划,原话——「有方向计划了,但是没有工程计划,没有具体的任务安排和验收清单标准……应该是我在任何一个 session 加载都能马上按照计划内的具体任务和台账去干活。你这个太初级了。」重做的 v2 三件套(总览 HTML + 事实索引 md + 工单集 md,见 §已知参照)被认可。本 skill 把 owner 这次亲口验收过的形态固化成可复用规程,今后任何项目出「编排计划」不许再退化成只有方向的初级版。
0 · 使用时机
- owner 或项目制度要求产出「编排计划」「排期」「任务安排」「工单包」类交付物时。
- 已出的计划被打回、理由类似"看不出具体怎么干/验收标准是什么/换个人接得上吗"时,直接重做,不必等第二次打回。
- 不适用:单张任务卡、一次性小修复、纯调研报告——那些不需要双层结构,硬套等于过度工程。判断线同
feature-dev的规模定档:改动面单一、无需多执行体协同、无跨阶段依赖图,就不必发本 skill,直接干或写一张卡即可。
1 · 双层结构(合格判据的核心,一条不能丢)
编排计划必须同时有两层,只有一层=不合格,直接打回自己重做,不要呈给 owner:
- 方向层:目标态是什么、为什么现在做、分几个大阶段、大阶段之间的因果关系。这层回答"我们要去哪、为什么"。
- 工程层:方向层拆到底之后,逐条落地的工单——每条工单必须自带执行体/依赖/输入/步骤/产出/可判定验收标准六要素(§3 给固定模板)。这层回答"具体谁、干什么、怎么算干完"。
自检一句话:拿掉方向层,工程层能不能独立执行?拿掉工程层,方向层能不能指导任何一个具体动作?两者都答"不能"才算双层齐全。只写方向层=书生做的规划,owner 原话定性为"太初级"。
2 · 三件套产出规程
owner 钦定 HTML 为总览与确认层的格式,但允许 HTML 里套 md 索引避免 HTML 臃肿——所以标准产出永远是三个文件,分工固定,禁止把三件套压成一件套或反过来拆成四件以上:
| 文件 | 角色 | 内容 |
|---|---|---|
<前缀>-编排计划-<主题>-<日期>.html |
总览与确认层(owner 第一眼读的) | 老板版首节 + 任务总表 + 依赖图 + 逐站验收门 + 候 owner 清单;链接指向另两个 md |
<前缀>-关键事实索引-<日期>.md |
事实层(冷启动第一份读物) | 版本/环境/账号/权威规范文件路径/已定谳件清单/会话表/纪律判据——任何新 session 不用问人就能补齐上下文 |
<前缀>-工单集-<范围>-<日期>.md |
执行层(真正干活时打开的) | 逐任务工单,自包含六要素,禁止"同上"式偷懒到看不出边界(可以写"步骤/产出/验收同模板"但每条的输入与产出文件名必须唯一列出) |
产出顺序:先事实索引(把散落的事实收拢),再工单集(把方向拆成工程),最后 HTML(汇总前两者,写老板版首节)。反过来先写 HTML 空表格再回填,容易漏工程细节。
3 · 冷启动自包含判准(验收计划是否合格的唯一硬标准)
owner 验收编排计划只问一件事:任何一个不带任何背景的的新 session,只读三件套,能不能马上按工单去干活,不用回来问我们、也不用现挖代码史。 逐条自检:
- 工单里的路径是不是绝对路径、文件是否真实存在(不是"应该在"那种猜测路径)?
- 依赖的分支/tag/commit hash 是否写死(不是"当前分支",因为分支会漂)?
- 权限/账号/环境地址是否在事实索引里给全,而不是要求读者去问 owner 才知道?
- 每条工单的"产出"文件名是否和"验收"里引用的文件名对得上?
- 每条工单是否写了模型档与预算档?复核是否按「只改测试 / 读侧一两行 → 主编排亲核;产品码或库函数 → 派复核员(八项清单、四小时封顶)」分级?
任一条答不上=不合格,回 §1 补,不许呈给 owner。
4 · 「第一节 · 总结说人话版」铁律(HTML 第一节;20260902 owner 打回后升级,标题字段名固定)
HTML 的第一个 <h2> 固定命名为**「第一节 · 总结说人话版」**(不再叫"老板版"),内容必须完整遵守 /human-explain skill 规范(禁 AI 腔与管理黑话:本质上/闭环/链路/落地/对齐/抓手…全禁;禁比喻替代直说;结论先行;术语先说意义再给名字),面向不想读代号的决策者:
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 · 213 lines · 144 tokens per session scan A 7901d4c437d5
orch-plan is a skill published in the GitHub repository 0fuk/OpenACCP (2 stars, last pushed 2d ago), licensed MIT. It adds 144 tokens to every session and 5,687 once invoked, about $0.0006 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
coordinate-agents
Route requests that explicitly use Coordinate Agents or operate on its existing Tasks to setup, execution, review, or recovery skills.
company-handbook
The cross-director coordination protocol for the 1ai business kingdom. Defines how Content, Video, Image, Marketing, and Sales directors work together. Has memory via 1ai-hub brain, follows 1ai-rules, and coordinates with Vilona (GM). Use when coordinating multiple directors, resolving conflicts between teams, or when…
coordination-bus
Use when two or more agents share work, when a task spans sessions, or when claiming, heartbeating, completing, or blocking a coordination bus task. The bus is the single source of truth for who owns what — never coordinate through private agreement.
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…
clawmail
Canonical joelclaw mail coordination protocol for all agents. Use when announcing work, checking inboxes, reserving files, releasing locks, handling task handoffs, triaging stale reservations, or auditing coordination quality. Triggers on: 'clawmail', 'agent mail', 'joelclaw mail', 'reserve files', 'release locks'…
nexus
Use in repos coordinated by the Nexus CLI, especially when NEXUSCONSTITUTION.md, NEXUSQUEUE.md, NEXUSSTANDUP.md, or .nexus/ exists, or when the user asks about Nexus start/claim/release/doctor/status/next.