Borrowing it
Nothing to install: this file belongs to zxc663/shisan-xinuo-workflow. 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/zxc663/shisan-xinuo-workflow/main/AGENTS.mdgit clone --depth 1 https://github.com/zxc663/shisan-xinuo-workflowWrote 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/zxc663/shisan-xinuo-workflow/agents-md)<a href="https://agentmods.dev/instructions/zxc663/shisan-xinuo-workflow/agents-md"><img src="https://agentmods.dev/badge/instructions/zxc663/shisan-xinuo-workflow/agents-md/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/instructions/zxc663/shisan-xinuo-workflow/agents-md"><img src="https://agentmods.dev/badge/instructions/zxc663/shisan-xinuo-workflow/agents-md.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 | $0.01985 | $0.01985 |
| Opus 5.5 | $0.00794 | $0.00794 |
| Sonnet 5.5 | $0.00397 | $0.00397 |
| Haiku 4.5 | $0.00198 | $0.00198 |
Grade A, and why
shisan-xinuo-workflow AGENTS.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 14d 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 — 29 lines — stays where its author put it; the contents beside it link to each section on GitHub.
项目级 Agent 规则 · shisan-xinuo-workflow（本仓库）
本仓库 = 十三希诺 Agent 工作流 Skill 的源库（开发库）。本文件是项目级注入（每会话自动进入），与平台全局硬注入（injection-core，负责通用纪律）互补：项目级管「本仓库特有信息 + 项目承载 + 维护纪律」。内容精简，细节按需读 docs/project-info.md（导航）。
工作流在场（本仓库会话）
- 开工序列四步（与全局硬注入一致，SKILL §2.0，v2.6.0 简化）：①复述理解（无条件先行）②承载检查（扫描+形态判定+定承载根+建/补一气呵成，memory/agent-log.md 已建则增量补缺）③记忆对齐（读 agent-log 状态段一屏 + 按症状检索，平台原生记忆在场不重复预读）④判级速查+三问选道；命 L3（密钥/删除/迁移/发布/架构/超预算）先问。
- 本仓库默认不 git push：本地 commit + 备份优先，推送须用户明确批准。
- 改 Skill 产物后：全仓 grep 承载点口径（细则条数/类数/版本号/在场提示）→ 跑
scripts/verify-release.ps1（内容锚点〔含 L3 判级双源对验〕/hooks/版本/泄漏/正文净化/索引/事实对账/判据自测/叙述对账/双副本一致性 10 项门禁）→ 同步各副本（python scripts/syncer.py，WorkBuddy 侧加--dest；family 包清单=显式数组，新包须登记）。 - 细则引用统一完整前缀
details #N/细则 #N（禁裸 #N——与 GitHub issue 编号同形异义，假阳性 9/10 实证）。 - 判级/红线权威源顺序：
references/injection-core.md（每会话在场）→SKILL.md（可执行细节）→项目信息.md（决策史）→docs/project-info.md（导航索引）。
维护纪律（本 Skill 正文 vs 史料——开发本 Skill 时的写作规范）
- 本 Skill 交付物正文（skill/ 下 SKILL.md/references/templates）只写规则本身 + ≤1 句为什么；版本出处 / 拍板人 / 日期 / 路测轮次只入本仓库决策史（项目信息.md / CHANGELOG / task-log / EVIDENCE），不入 Skill 正文。
- details 条目尾部
*来源/晋升*字段与节首来源注记 = 豁免（双击晋升准入证据，史料属性合法）。 - 注入核心自包含：injection-core.md 内禁悬空跨文件指针（须自包含或写全文件名）。
- 门禁：verify-release E 项（常驻/模板面）+ 全仓 grep 承载点口径，改动后必跑。
- 新增机制条款不带版本号/拍板人署名（决策记录写项目信息.md §三）。
- 细则双库（2026-10-13 解冻令，取代 2026-10-10 转型期冻结；docs/transformation-plan-20261010.md §六解冻决议 + 设计档 docs/design-specs/v420-plan-unfreeze-decouple-privacy-20261013.md）：
details.md（条款账）零和 405 上限不变，新增须合并/替换旧条；踩坑教训分拣迁references/lessons.md（教训账）解冻不占 405，入库须带[域]/[态]字段+反论证三问，靠生命周期代谢（修复→退役→复活回指，flows「教训生命周期与上下文卫生工作流」）；判据 j2.7/聚合器 a1.2/盲评夹具=维护态（修 bug 可以，不加层）；新增机制条款仍不带版本号/拍板人署名。
项目承载（已就绪）
memory/：会话记忆——gitignore 本地承载，不随仓分发（一档制agent-log.md四区；旧五件套历史原件在memory/legacy-pre-v250/只读保留）。docs/project-info.md：项目导航六节（架构/目标/模块表/调研导航/参考资源/签章）。项目信息.md：决策与发布史（权威承载）。dist/、versions/personal-zh/、.trae/：本地/私有，不随仓分发或按 gitignore 处理。
本仓库底线（区别于通用纪律）
- 密钥/令牌绝不写入本仓库任何文件（verify-release 泄漏红线 D 项会拦）；机密文档仅存本机专用机密目录（位置不在此写出、不随仓，以 memory 最新记录为准）。
- 发行动作（npm / GitHub Release / Gitee / ClawHub / About）必须先获用户批准 +
verify-release10/10 PASS + 观测阶段。 - 当前基线 = v4.2.0 批（细则双库解冻〔条款账零和 405 + 教训账解冻不占 405〕+教训域/态字段与生命周期协议〔active/fixed/retired，修复三要件，复活回指〕+普查批 A/B 三裁决落档〔域：通用 329/平台 77/环境 0〕+开源隐私门〔四层扫描 privacy_gate.py〕+时间戳强制场景+代拟用户言论词表机检面；上一发行态 v4.1.0 全渠道 2026-10-12，回执见 RELEASE-CHECKLIST P 节与项目信息 §五）；上一发行态 v4.0.0 全渠道 2026-10-10，回执见 RELEASE-CHECKLIST O 节；v4.1.0 批=门禁防线+部署链+规则层定型（Phase 1-8 施工：#103 真删 406→405/W8 卖点重述/j2.7 GATE 收敛循环到绿/注释纪律/单文件重蒸馏基线 v4.1.0/对外素材〔deepswe+A-4〕/审计小件〔聚合器 a1.2 mismatch 不可比+probe_runner 退出码契约+syncer F-10/F-11+scorecard 双指标口径〕；上一发行态 v4.0.0 全渠道 2026-10-10，回执见 RELEASE-CHECKLIST O 节与项目信息 §五）**——细则 405 条·33 类（编号至
#407）+ 判据 j2.7（金样本 25/25；GATE 收敛语义「循环到绿」，JUDGELOG 变更 12）+ 门禁 11 项（A-K：内容锚点含 L3 判级双源对验 / hooks / 版本 / 泄漏 / 正文净化 / 索引 / 事实对账 / 判据自测 / 叙述对账 / 双副本一致性〔CI 无分发根=显式 SKIP〕/ 开源隐私门 privacy_gate 四层扫描〔-SkipPrivacy仅结构校验，发行跑全量〕）+ 五个机检端口（risk_scan/agent_log_rotate/gate_audit --gate/--high-risk/--independent-cmd/deploy_injection --check --hash/privacy_gate --check/--report/--selftest）+ 叙述对账机制narrative_sync（F8，已接线 verify I 项）。三面校准以 sync-all 跑后回执为准。新会话验收锚=「在场提示」版本行 +405 条细则+zxc663应答（注入版本=会话创建时快照，GUI 长活会话须重启应用）。
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.
- 14d ago Changed · +315 tokens per session 337465a3a7b5
- 16d ago Changed · +57 tokens per session e15786e68592
- 17d ago Changed · +153 tokens per session 45f7b769badb
- 18d ago Changed · +1 lines · +163 tokens per session 49f7f321f431
- 21d ago Changed · +14 tokens per session bf569398c0b9
- 28d ago Changed · -40 tokens per session 71b6c6b2ffdc
- 29d ago Changed · -38 tokens per session 0457a7e9d296
- 1mo ago Changed · +80 tokens per session 99b9930d724c
- 1mo ago Changed · +27 tokens per session 9b1c244a065a
- 1mo ago First seen · 28 lines · 1,254 tokens per session scan A 9c621c687941
shisan-xinuo-workflow AGENTS.md is an instructions file published in the GitHub repository zxc663/shisan-xinuo-workflow (43 stars, last pushed 15d ago), licensed MIT. It adds 1,985 tokens to every session, about $0.0079 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 2026-08-30.
Other instructions, from other repositories
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.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
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 copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).
vscode model-prompts.instructions.md
Model-specific prompt authoring and registry guidelines.