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 lawyerllsx-dotcom/Legal-Workflow-Chain --skill legal-weekly-briefinggit clone --depth 1 https://github.com/lawyerllsx-dotcom/Legal-Workflow-ChainWrote 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/lawyerllsx-dotcom/legal-workflow-chain/legal-weekly-briefing)<a href="https://agentmods.dev/skills/lawyerllsx-dotcom/legal-workflow-chain/legal-weekly-briefing"><img src="https://agentmods.dev/badge/skills/lawyerllsx-dotcom/legal-workflow-chain/legal-weekly-briefing/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/lawyerllsx-dotcom/legal-workflow-chain/legal-weekly-briefing"><img src="https://agentmods.dev/badge/skills/lawyerllsx-dotcom/legal-workflow-chain/legal-weekly-briefing.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 | $0.00242 | $0.07768 |
| Opus 5 | $0.00121 | $0.03884 |
| Sonnet 5 | $0.00048 | $0.01554 |
| Haiku 4.5 | $0.00024 | $0.00777 |
Grade A, and why
法律周报 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 yesterday.
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 — 398 lines — stays where its author put it; the contents beside it link to each section on GitHub.
法律周报自动化 · Legal Weekly Briefing
全局禁令（Agent 铁律）
以下规则在整个 skill 执行期间必须遵守。违反任一条 = 交付阻断。
| 场景 | 必须遵守 | 不应发生 |
|---|---|---|
| 入库知识库 | 仅写入本地 Obsidian weekly-briefing-archive/notes/ 目录，不经过任何第三方服务器 |
不应指引订阅/加入/发送到任何外部知识库 |
| HTML 交付 | 仅用 render_html.py 渲染（#f8f7f5 / #1a1a2e 浅色简报风） |
不应自造深色翻页幻灯片或其他替代样式 |
| 交付验证 | 每次修改后必须跑 python3 scripts/verify.py，全量通过 |
不应跳过验证或仅凭人眼判断 |
| 候选数据 | 每条候选必须含 abstract + recommend |
不应构建缺少必填字段的 candidates.jsonl |
| session.json | 仅本地使用，.gitignore 已排除 |
不应分享/提交到 Git/发送到任何服务器 |
| 样式修改 | 需要改样式时先问用户确认 | 不应擅自改动配色/布局/交互 |
| 适配流程 | 用户说「配置法律周报」→ 必须按适配向导 4 问引导 | 不应跳过适配向导给通用配置 |
| 内容发现路径 | 启动时检测双通道（微信读书登录态 → 元宝登录态），确定路径后明确告知用户 | 不应静默降级到 WebSearch 而不告知路径变化及影响 |
| 候选摘要 | 每条候选 abstract 必须非空，三级回退见 references/curation-guide.md |
不应产出 abstract 为空的候选条目 |
工作流
Step 0 · 启动前检查更新（必做）
每次启动本 Skill 前，先在 Skill 根目录检查 GitHub 上游是否有更新；有更新时先问用户是否要更新，用户确认后再执行更新，然后继续后续流程。
git -C "<SKILL_ROOT>" fetch --quiet
git -C "<SKILL_ROOT>" rev-list --count HEAD..@{u}
如果返回值大于 0，告诉用户检测到上游更新数量，询问是否先执行：
git -C "<SKILL_ROOT>" pull --ff-only
执行规则：
- 不要自动更新。用户拒绝时继续使用当前版本。
- 如果网络不可用、没有 upstream、不是 git 仓库、或 remote 被 404/删除，静默跳过本步，直接进 Phase 0，不报错不打断。
- 当前状态：remote 已配置为 https://github.com/5tnb6xgsm5-ops/legal-weekly-briefing，Step 0 已生效。
Step 1 · 进入 Phase 流程
完成 Step 0 后，按下方「标准执行流程（Phase 0-6）」执行。
第一性原理
同一批法院公众号文章，走三层分流：
Tier 1 · 精读区：k-NN 评分引擎挑 12 条精品（AI+法律 5 + 纯法律 7）→ 律师主动阅读 Tier 2 · 雷达区 + Obsidian 归档：低分≠消失，legal 未进精读的前 8 条进「其他领域速览」（雷达区）防闭门造车；score ≥ 阈值法院源条目全量写入本地 Obsidian 归档目录（weekly-briefing-archive/notes/） Tier 3 · 噪音：非 legal 源 + 非 court 源 + 未进 AI 精读的条目，自然落选
精读解决"本周重点看什么"，雷达区解决"执业圈外还有什么在动"，Obsidian 入库解决"以后能找到什么"。三层共享内容发现层，在评分环节分叉。
核心交付模式：「配置一次，每周自动推送」——依赖外部调度层（WorkBuddy Automation / GitHub Actions cron）定时触发。
分级架构
Level 0 · 5 分钟快速体验（零配置，零依赖）
└─ 预置 10 条示范候选 → demo.py → 演示周报 MD+HTML
Level 1 · 纯评分引擎（零外部依赖）
└─ 用户提供候选 URL 列表 → k-NN 评分排序 → 周报 MD+HTML
Level 2 · + Obsidian 本地入库（零外部依赖）
└─ Level 1 + obsidian_importer.py → 分类 → 写归档笔记（本地 weekly-briefing-archive/notes/）
Level 3 · + 自动发现（需微信读书账号 + 元宝账号）
└─ 微信读书搜一搜（主）→ 元宝反查（补）双通道拉取四账号文章
What ships with it
39 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- assets/config/settings.yaml 4.0 KB
- assets/config/sources.yaml 2.1 KB
- assets/config/taxonomy.yaml 4.3 KB
- assets/data/scoring-training.jsonl 107 KB
- assets/data/test-prompts.json 2.5 KB
- README.md 12 KB
- references/_DEPRECATED_mp-setup-guide.md 7.7 KB
- references/adaptation-wizard.md 8.1 KB
- references/automation-setup.md 1.0 KB
- references/candidates-schema.md 3.5 KB
- references/curation-guide.md 6.1 KB
- references/delivery-gate.md 5.1 KB
- references/feature-guide.md 19 KB
- references/obsidian-import-guide.md 4.2 KB
- references/weread-setup-guide.md 2.9 KB
- references/yuanbao-setup-guide.md 3.2 KB
- references/全方位测试报告-2026-08-01.md 3.9 KB
- references/新渠道内容质量解决方案.md 5.0 KB
- references/评分体系维护指南.md 4.6 KB
- scripts/build_candidates.py 18 KB runs code
- scripts/dedupe.py 2.4 KB runs code
- scripts/demo.py 15 KB runs code
- scripts/fetch_weread_week.py 9.2 KB runs code
- scripts/fetch_yuanbao_supplement.py 12 KB runs code
- scripts/install.sh 2.5 KB runs code
- scripts/merge_candidates.py 3.8 KB runs code
- scripts/normalize_url.py 1.7 KB runs code
- scripts/obsidian_importer.py 8.6 KB runs code
- scripts/publish_check.sh 3.3 KB runs code
- scripts/render_html.py 20 KB runs code
- scripts/run_pipeline.py 30 KB runs code
- scripts/scoring_engine.py 13 KB runs code
- scripts/validate-weekly-output.mjs 15 KB runs code
- scripts/verify.py 14 KB runs code
- scripts/weread_login.py 4.2 KB runs code
- scripts/wiki_ingest.py 9.1 KB runs code
- scripts/yuanbao_login.py 4.7 KB runs code
- 周报_demo_2026-09-09.html 28 KB
- 周报_demo_2026-09-09.md 4.8 KB
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.
- yesterday First seen · 398 lines · 242 tokens per session scan A 09ed1303b712
法律周报 is a skill published in the GitHub repository lawyerllsx-dotcom/Legal-Workflow-Chain (5 stars, last pushed 2d ago), licensed MIT. It adds 242 tokens to every session and 7,768 once invoked, about $0.0012 per session on Opus 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 2026-09-11.
Other skills, from other repositories
tw-legal-rag
A tool for finding real Taiwan court judgments and legal interpretations with verifiable citations before answering questions about Taiwan law.
legal-research
A legal research skill for finding and analysing laws and court cases, including similar cases. It requires access to a reliable legal research database before producing a formal report.
document-fill
A document-filling assistant that uses a supplied template and case files to fill in forms, contracts, or legal documents. It also reports where each filled value came from and which details are missing.
contract-review
A contract-review workflow for agreements governed by Chinese law, United States law, or both. It examines clauses one by one and produces a list of legal-risk questions, with an optional marked-up Word document.
legal-kb-builder
A tool for turning legal documents and other source material into a searchable local knowledge base. It supports sources such as PDF, DOCX, images, Markdown, and question-and-answer sets.
kr-patent-consistency-check
A Korean patent-document checker for comparing a patent specification, drawings, and claims—the sections that describe an invention and define its legal protection.