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 open-octo/octo-agent --skill legal-due-diligencegit clone --depth 1 https://github.com/open-octo/octo-agentWrote 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/open-octo/octo-agent/legal-due-diligence)<a href="https://agentmods.dev/skills/open-octo/octo-agent/legal-due-diligence"><img src="https://agentmods.dev/badge/skills/open-octo/octo-agent/legal-due-diligence/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/open-octo/octo-agent/legal-due-diligence"><img src="https://agentmods.dev/badge/skills/open-octo/octo-agent/legal-due-diligence.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00155 | $0.03533 |
| Opus 5 | $0.00077 | $0.01767 |
| Sonnet 5 | $0.00031 | $0.00707 |
| Haiku 4.5 | $0.00015 | $0.00353 |
Grade A, and why
legal-due-diligence 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 8d 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 — 211 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Skill: legal-due-diligence
面对一批文档（合同、公司档案、诉讼记录等）需要系统性审查时，有两种互补的做法：
- 表格式（tabular）：同样一组字段问遍每一份文档，产出一行一文档、一列一 字段的表格。适合"这 50 份合同里，变更控制条款分别怎么约定的"。
- 问题提取式（issue extraction）：按标准类别清单扫描，只把真正有问题的 文档挑出来写成发现清单。适合"这批材料里有没有埋雷"。
两者可以先后使用：先跑表格式摸清全貌，再对表格里标红/标黄的行做问题提取式深挖。
这不是替代人工阅读文档。 每一格/每一条结论都是"线索"，需要人工核实，不是 "定论"。这个技能的目标是让核实变快，不是让核实变得可以跳过。
处理规模的现实约束
当前 profile 没有配置并行子代理（sub_agent），本技能按文档顺序逐份处理， 不做原始方法论里"每份文档一个并行子代理"的 fan-out。文档量较大（50+）时提前 告知用户这会比较耗时，并建议先用小样本（3-5 份）跑通字段定义再处理全量，避免 字段定义有问题却已经审完几十份文档。
逐字引用铁律
这是整个技能最重要的纪律，不是可选项：
每一条结论（不管是表格里的一个格，还是问题清单里的一条发现）必须配一段来自 原文、逐字复制的引用，以及能让人重新定位到原文的位置信息（章节号/条款号/页码， 文档给了什么就用什么）。
- 不能把一个标题加常见样板文字拼成"引用"。
- 不能改写后当作逐字引用。
- 不能凭"这类条款通常怎么写"的印象复原一段引用。
- 找不到原文、定位不到，就把这一条标记为"待核实"（见下面的三态规则），value 留空，并在备注里写清楚原因（文档截断、扫描件识别不清、条款隐含但没写明、 只看到标题没看到正文等）。绝不能为了让格子显得"已完成"而编一个引用。
这条纪律同样适用于所有字段类型的"配套原文引用"，不只是"逐字类"字段本身—— 一个分类判断（比如"需经同意"）如果配的引用是编的，比留空更危险，因为它看起来 更完整、更容易被直接采信。
三态"未找到"规则
一个空格子会掩盖信息。凡是给不出正面答案的情况，强制归入以下三种明确状态之一：
| 状态 | 含义 | 使用场景 |
|---|---|---|
未涉及 |
读过文档，确认没有这条约定 | 有把握该主题文档确实没提 |
不确定 |
文档里有相关内容，但无法确信地分类 | 措辞含糊、条款不完整、内容自相矛盾 |
待人工判断 |
找到了内容，但需要人判断怎么归类 | 边缘情形、罕见措辞、答案取决于字段定义没覆盖的判断 |
"合同对此完全没有约定"和"约定含糊看不清楚"是团队会用完全不同方式处理的两种 情况，压缩成一个空格子会丢失这个区别。
模式一：表格式提取
第一步：定义字段和范围
和用户确认：审哪些文档（本地文件/用户粘贴的文本）、要哪些字段、输出放哪里。
把用户的字段描述整理成结构化 schema，每个字段包含：一个稳定的 id、一个人类
可读的 名称、一个 类型、一个 提示词（一个真正在读文档的人会问的问题），
分类型字段还需要一个 选项列表：
| 类型 | 返回什么 | 用于 |
|---|---|---|
逐字 |
原文精确引用，一字不改 | 定义性术语、操作性条款原文、任何措辞本身很关键的地方 |
分类 |
从你预定义的固定选项里选一个 | 是/否、有/无、条款变体（如"需经同意"/"同意不得无理拒绝"/"未提及"） |
日期 |
ISO 日期 | 生效日、到期日、终止通知截止日 |
期限 |
数字+单位 | 合同期限、通知期、存续期 |
金额 |
数字+币种 | 责任上限、门槛、费用 |
数字 |
裸数字 | 数量、百分比、页码引用 |
自由文本 |
简短自由文本摘要 | 谨慎使用——这是最容易漂移的类型，其他类型确实不合适时才用 |
逐字引用规则同样适用于非"逐字"类型的字段：每个非逐字字段都要配一个"支持 原文"作为配套字段。格子里的答案是解读，配套引用是证据——一个写着"同意不得无 理拒绝"的分类结果，没有对应的原文句子支撑就没有意义，因为核实者的工作就是检 查这个解读对不对。
用小样本（3-5 份文档）先跑一遍，检查：某个字段是不是大部分答案都是"不确定" （说明提示词本身模糊，需要改写）、"分类"字段的答案是不是经常不落在预设选项里 （需要补充选项或改成自由文本）、"逐字"字段是不是经常返回的是改写而非原文（需要 在提示里再强调一次"必须逐字"）。调整后再确认，避免整批跑完才发现字段定义有问题。
What ships with it
2 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.
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.
- 8d ago First seen · 211 lines · 155 tokens per session scan A 32aaf35cffcb
legal-due-diligence is a skill published in the GitHub repository open-octo/octo-agent (99 stars, last pushed yesterday), licensed MIT. It adds 155 tokens to every session and 3,533 once invoked, about $0.0008 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 2026-09-03.
Other skills, from other repositories
uspto-database
Access USPTO APIs for patent/trademark searches, examination history (PEDS), assignments, citations, office actions, TSDR, for IP analysis and prior art searches.
fda-database
Query openFDA API for drugs, devices, adverse events, recalls, regulatory submissions (510k, PMA), substance identification (UNII), for FDA regulatory data analysis and safety research.
ha-mac-control
Hope Agent native macOS desktop control — the standard maccontrol status / diagnostics / apps / dock / spaces / snapshot / visual / windows / menu / clipboard / dialog loop, target-first action rules, no-blind-coordinate policy, and recovery for stale AX/window/menu/dialog state. Load whenever using maccontrol, or…
ha-skill-creator
Create, edit, improve, or audit Hope Agent skills. Use when the user wants to: (1) create a new skill from scratch, (2) edit or improve an existing skill, (3) review or clean up a SKILL.md file, (4) run evaluations to test skill effectiveness, (5) optimize skill descriptions for better trigger accuracy. Trigger…
ha-browser
Hope Agent browser automation — the standard status → tabs → snapshot → act loop, stale-ref recovery rules, and what to do when login / 2FA / captcha / camera-prompt / dialog blocks progress. Load this skill whenever you reach for the browser tool. Trigger on: user asks the agent to open / control / click / scrape /…
ha-logs
A read-only troubleshooting skill for querying Hope Agent’s local SQLite databases, which store logs, conversations, and background-job status.