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 5gategit 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/5gate)<a href="https://agentmods.dev/skills/0fuk/openaccp/5gate"><img src="https://agentmods.dev/badge/skills/0fuk/openaccp/5gate/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/5gate"><img src="https://agentmods.dev/badge/skills/0fuk/openaccp/5gate.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 | $0.00182 | $0.05319 |
| Opus 5.5 | $0.00073 | $0.02128 |
| Sonnet 5.5 | $0.00036 | $0.01064 |
| Haiku 4.5 | $0.00018 | $0.00532 |
Grade A, and why
5gate 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 — 106 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/5gate — 五门发版范式
一句话:一个问题从「要修」到「在生产上被真人用过」,固定经过五道关,每道关只认落盘的证据,不认口头声称;哪一门空着,后面的门就不许进。 项目细节(路径、账号、部署脚本)读项目根 PROJECT-PROFILE.md;人读版宪法放在项目规范目录(路径登记在项目档案),与本 skill 冲突以宪法为准、改必须两处同改。
第 0 步 · 开车前(不算门)
按 /orch-plan 出本车编排计划三件套;工单集里每张卡自带「五门格」(五个格子,只许填证据文件路径)——这张工单集就是本车的五门台账,不另开账本。新问题一律先在台账立卡再动手;没有卡的改动=乱发明,当场停。owner 在此说「确认开工」。
五门(每门:谁 / 进门条件 / 出门判断标准 / 产物)
| 门 | 谁 | 进门条件 | 出门判断标准 | 产物(填进五门格) |
|---|---|---|---|---|
| ①任务门 | 各修复线,各自 worktree | 有「复现出错」的证据(已有盲测红卡/取证件的直接引用;只有零证据的新立项才派一个复现代理先造红) | 修复线的红绿验证会抓错(先故意改坏证明测试会变红)+ 不带背景的独立复核代理复核 + 关键宣称回数据库/真机核对 | 修复交回件 + 复核件 |
| ②合流门 | 主编排 | 本车全部卡的任务门都齐 | 候选 tag 五处来源互相核对 + 期望 commit 清单 + 封版前三检(动过被封印的源必重封 / 写死「当前尾条」的测试同步 / 跑测试脚本健康) + 全量回归零新增红(两次候选 tag 之间差异与某面测试零交集才许沿用结果,判读件贴 diff --stat) | tag + 全量交回件 |
| ③盲测门 | 部署线换装 + 盲测员 | 测试数据像真的(与生产同形态);章程含验法栏且 tools/blind-check.mjs 过;staging 已换到候选 tag |
卡组按类型分组(资金/权限/审批/展示…),同一 staging、同一批账号挨个跑,只走界面(允许被动抓包与只读探针取证);关键宣称回数据库核对;主编排三路交叉判读「阻发车项=0」 | 章程 + 盲测交回 + 判读件 |
| ④发车门 | 部署线 | 判读件落盘,授权口径已在(项目档案记 owner 常设授权则不再等) | 推远端(tag+分支)→ 发车件(一份三节:判读/给 owner 与门店的说明/给部署线的执行令)→ 部署四门(重封+一致性检查 / 构建物来源核对+commit 清单 / 盲测 PASS 判读件 / 构建物在 tag 现建)+ 规范备份门(规范/规则类文档镜像仓同步到最新并打与生产同名 tag 推远端;项目档案记仓址)+ 备份真恢复演练 → 回滚窗四项判断 | 发车件 + 发车交回 |
| ⑤真人门 | owner 安排的真人 | 生产已切换 | 真人走完核心流程;红 → 回①立项(下一车或热修),不在生产边测边改 | 反馈件 |
门次触发表(主编排照表走,不靠想起来;owner 不需要手动调 verify-3way)
- 修复交回件落盘 → 按 verify-3way §4 做①门出门三件。
- 台账里所有卡①门齐 → 进②门:出候选 tag,并行派全量回归与 staging 换装。
- 换装交回落盘 → 按 verify-3way §2 出分组盲测章程、派盲测员。
- 盲测交回落盘 → 按 verify-3way §3 出交叉判读;阻发车项=0 → 进④门发发车件。
- 发车交回落盘 → 登记(台账/PROFILE/memory/修复池收编)→ ⑤门候真人反馈。
两条回环规则
- ③门出红:只对那张红回①(从当前候选 tag 拉新 worktree 修)→①→②出新候选 tag(全量按「可否沿用」判定决定重跑面)→换装→只重放红卡+抽验回归,不整章重跑。
- 途中新发现:一律进修复池立卡,不进本车;owner 亲口「进本车」才装。
并行轴(别串行浪费)
①门各线并行;②门全量与③门换装并行;③门盲测一员顺序走(见用量纪律;只有两份章程互不相干且账号与数据不交叉时才分两员)。
发车件模板(三节合一,取代判读/发车说明/发车令三份文件)
文件名 <前缀>-发车件-<版本>-<日期HHMM>.md;第一行必须是「台账 <工单集绝对路径>:全部 N 卡①②③门齐,阻发车项=0」。
- 第一节·判读:候选 tag 与 SHA;卡组盲测结果三栏(绿/红/灰各几张,红为零才有本节);全量回归读数与「可否沿用」判定;阻发车项=0 的依据文件路径。
- 第二节·说明(给 owner 与门店):说人话——这车修了什么、门店会看到什么变化、需要谁做什么(运营待办逐条);不出现代号。
- 第三节·执行令(给部署线):基座 SHA/tag、期望 commit 清单、迁移号区间与预期台账数、部署四门逐项判据、备份与真恢复演练要求、回滚窗四项判断与时长、交回件文件名。
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 · 106 lines · 182 tokens per session scan A 43f40244110b
5gate is a skill published in the GitHub repository 0fuk/OpenACCP (2 stars, last pushed 2d ago), licensed MIT. It adds 182 tokens to every session and 5,319 once invoked, about $0.0007 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
Agentic Orchestration for Oracle
Multi-agent coordination patterns for enterprise AI systems on Oracle Cloud Infrastructure.
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…
api-workers
Cloudflare Workers deployment using createWorkerHandler from @cyanheads/mcp-ts-core/worker. Covers the full handler signature, binding types, CloudflareBindings extensibility, runtime compatibility guards, and wrangler.toml requirements.
vercel-deploy
Deploy applications and websites to Vercel. Use this skill when the user requests deployment actions such as "Deploy my app", "Deploy this to production", "Create a preview deployment", "Deploy and give me the link", or "Push this live". No authentication required - returns preview URL and claimable deployment link.
architecture-diagram
Dark-themed SVG architecture/cloud/infra diagrams as HTML.
moai-platform-database-cloud
Cloud database platform specialist covering Neon (serverless PostgreSQL), Supabase (PostgreSQL 16 with real-time), and Firebase Firestore (NoSQL with offline sync). Use when choosing or setting up cloud databases.