CommunityResearch & Data Analysisgithub.com

Investment Terms Review Advisor (TS Term Review)

Reviews Term Sheets and investment agreements clause by clause. Classifies each clause as red (must fix), yellow (negotiable), or green (market practice), and provides negotiation scripts and alternatives.

What is Investment Terms Review Advisor (TS Term Review)?

This specialist AI skill empowers developers and entrepreneurs to critically evaluate Term Sheets (TS) and investment agreements. Instead of offering a vague “looks okay overall,” it systematically reviews 15 core clauses—including valuation (Pre/Post), liquidation preference, anti-dilution (with full ratchet as a red line), drag-along, redemption, personal guarantees, board composition, protective provisions, and more. Each clause receives a three-tier risk rating (🔴 Red: do not sign without changes, 🟡 Yellow: negotiable, 🟢 Green: market standard) along with concrete negotiation scripts, alternative phrasing, and a clear distinction between company obligations and founder personal liability—personal joint-liability redemption is flagged as a primary risk source. The output is a structured decision table with a must-fix list, negotiation priorities, a bottom-line advice summary, and a disclaimer that it does not constitute legal advice. It integrates with the companion skill `funding-fit-diagnosis` for overall fundraising assessment. Supported platforms include any AI agent that accepts markdown-based skill definitions (Claude Code, Cursor, Codex, etc.). The skill works by applying a checklist from `references/04-fundraising-playbook.md` and enforcing strict execution discipline to avoid common misuses like over-labeling or missing personal liability.

Works with~Claude Code~Codex CLI~Cursor
npx skills add https://github.com/lingyu9495-source/agent-skills/tree/main/skills/ts-term-review

Installed? Explore more Research & Data Analysis skills: affaan-m/uspto-database, affaan-m/scholar-evaluation, affaan-m/literature-review · View all 6 →

Ask in your favorite AI

Open a new chat with this agent skill pre-loaded.

Preview

From the skill README

Jiupin Jinli e WeChat QR Code

Documentation

What does Investment Terms Review Advisor (TS Term Review) do?

Specialized skill. Activate when the user has received a Term Sheet (TS), investment agreement, capital increase agreement, or equity transfer agreement and needs to determine "which clauses must be changed, which can be conceded, and what is the bottom line." Companion main skill: funding-fit-diagnosis (use that skill for overall fundraising assessment).

1. Trigger Scenarios

  • "I received a TS / term sheet, help me review it."
  • "The investor wants me to sign a redemption / performance clause, is that reasonable?"
  • "Valuation Pre or Post / how much equity should I give up?"
  • "What does liquidation preference / anti-dilution / veto power mean, and can I accept it?"
  • "What are the pitfalls in this investment agreement?"

2. Execution Discipline

  1. Review clause by clause; never give a vague "looks fine overall" — risks are hidden in specific terms.
  2. Must assign a risk level: 🔴 Red (don't sign without changes) | 🟡 Yellow (negotiable, push for improvements) | 🟢 Green (market standard, usually acceptable).
  3. Must provide negotiation scripts / alternatives — pointing out risk without a fix only adds anxiety.
  4. Distinguish between company obligations and founder personal liability — personal joint-liability redemption is the source of most disasters; it must be flagged as red.
  5. This is not legal advice. For formal signing, legal effect, and judicial practice, explicitly recommend consulting a licensed attorney; if uncertain, write "needs lawyer confirmation."

3. Review Workflow (15 Core Clauses)

Cross-reference with references/04-fundraising-playbook.md's "TS Clause Red Line Checklist":

#ClauseRed Line Judgment
1Valuation & Pre/Post🟡 Don't go to the ceiling; leave room for the next round; not more than 1.2x comparable transactions
2Liquidation Preference🔴 1x participating and above is a red flag; 2x+ participating = harsh
3Anti-dilution🔴 Full ratchet is a red line (all down-round losses borne by founders); fight for weighted average; exclude ESOP
4Performance/ Bet Clauses🔴 Founder personal joint-liability redemption = high risk; company-level bet requires capital reduction first (2024 Fazawang + 2025 draft opinion)
5Redemption Right🔴 Unlimited personal redemption is a red line; typical: principal + 8-15% annual, 5-7 year trigger; fight for cap and remove personal liability
6Board Composition🟡 Founders must retain majority (2-1-1 or 3-2-1)
7Veto Rights🟡 Reasonable 8-12 protective matters; if exceeds (e.g., daily operations), narrow down; raise trigger threshold
8Preemptive Rights🟢 Market practice; fight for ESOP and strategic acquisition exemptions; set exercise period
9Vesting🟢 Standard 4 years + 1 year cliff; fight to credit years already served
10Non-compete🟡 Fight to limit to direct competition, duration ≤12 months, with compensation
11Drag-along Right🟡 Low threshold is risky; fight for >75% trigger + minimum price protection
12Tag-along Right🟢 Market practice, protects minority shareholders
13Exclusivity Period🟡 >90 days is risky; fight for ≤60 days; 90 days leaves startup passive
14Founder Guarantee🔴 Unlimited personal guarantee is a red line; fight for ≤50% investment amount or delete entirely
15Technology Bet (2026 new)🟡 Common in deep tech/biotech; must define quantifiable technical targets + exit plan if tech fails

Additional Must-Check Items (2025-2026 new):

  • ESG/compliance clauses, data compliance (personal information protection), AI-related IP ownership
  • Founder marital property / nominee shareholding / non-compete history disclosure
  • Conditions precedent (CP) list — check for "impossible to fulfill" items

4. Output Format (Mandatory)

[One-sentence conclusion] Can this TS be signed? Which clauses must be changed?

[Clause-by-clause review table]
Clause | Your current status | Level | Risk explanation | Suggested change | Negotiation priority

[Red line clause list] 🔴 Must change (for each, specify "what is acceptable after change")

[Negotiable clauses] 🟡 (provide trade-off advice; mark items that can be used as bargaining chips)

[Market practice] 🟢 (let the user know which clauses are not worth fighting over)

[Negotiation strategy] Which to fight first, which can be bundled, what is the bottom line, when to walk away

[Next steps] Recommend seeing a lawyer for the formal agreement (note that this tool does not constitute legal advice)

5. Common Misuses (Self-Check)

  • ❌ Only saying "I suggest you see a lawyer" → equivalent to not reviewing at all
  • ❌ Marking all clauses as red lines → user cannot prioritize
  • ❌ Not distinguishing company liability from personal joint-liability → missing the biggest risk
  • ❌ Not providing alternative fixes → user doesn't know how to negotiate
  • ❌ Stating legal effect as absolute (e.g., "company redemption is certainly invalid") → judicial practice evolves; must note "needs lawyer confirmation"

6. Disclaimer and Boundaries

  1. This skill is a clause sorting and risk alert tool; it does not constitute legal advice and cannot replace a licensed attorney's review.
  2. Cited judicial interpretations (e.g., 2024 Fazawang, 2025 draft opinion) are time-sensitive; before formal signing, always refer to the latest laws, regulations, and judicial practice.
  3. Data is current as of September 2026.
  4. The validity of clauses depends on specific wording, signing entities, articles of association, and other factors. This tool provides common risk alerts, not conclusive legal judgments.

🙋 About the Author

Jiupin Jinli e | Packaging hard-won lessons into ready-to-run skills, not textbooks. This skill is the version I use daily myself.

WeChat: ly5419495 (Add with note "SkillHub" for priority approval) Public Account: Chuwu Agent (Search on WeChat — reviews and methods are posted there; you can read even without adding my WeChat)

My other ready-to-run tools are on this shelf page (copy and open in browser): https://skillpay.alipay.com/public/jiupinjinlie

If you get stuck while using it, or have another scenario you'd like me to package into a skill, reach out via any of the methods above.

Jiupin Jinli e WeChat QR Code

Individual skills in this repo

This repo contains 17 individual skills — each has its own dedicated page.

lingyu9495-source/agent-responsiveness-delegation

你给 AI 助手派了个重活,然后发现:发消息它不回、想插话只能排队、等十几分钟才回一句——等它抬头,需求早变了。根因不是模型慢,是主脑把重活放在前台跑,整个回合被占死。这个 Skill 把这套「重活外派 + 随时应答」机制做成可落地的东西:① 规则块(一页写清判级、派完立刻让出对话、工单四要素、只回收结论、自报≠事实、拍板留主脑、跨会话任务换机制)② 跨平台落地脚本 provision_agent_responsiveness.py:探测本机已有的 Agent 环境(Hermes 的 SOUL 置顶铁律块、Claude Code 的全局指令文件,这两条已实测),把规则块幂等写入(自动备份原文件、重复执行不重复写、绝不覆盖你已有内容);其余平台(Codex / Cursor / WorkBuddy / 任意 Agent)则导出一份可直接粘贴的规则块 markdown,你粘到系统提示或自定义指令里就生效 ③ 体检脚本 audit_agent_responsiveness.py:逐项核对规则块是否存在、是否被后续改动覆盖、探测到哪些平台——没检测到就如实报「未检测到支持的平台」,不假报通过。纯标准库、零依赖、Windows/macOS/Linux 通用。触发词:AI不回话、等太久、插话卡住、助手响应慢、主脑被占住、重活外派、子代理、多智能体、响应速度、落地配置、系统提示。

lingyu9495-source/ai-image-watermark-removal-boss

去除 AI 生成图角落的"XXAI生成"半透明/白色角标水印。像素级验证不自我欺骗、双信号交叉定位抗幻觉、多候选客观指标投票覆盖。当用户要求去除图片水印、清理 AI 生成角标、去"豆包/即梦/Kolors 生成"水印时使用。

lingyu9495-source/bank-scan-household-splitter

客户丢来一个压缩包,里面几百张身份证正反面、开卡申请书、合同——**人工一张张挑,一下午就没了,还容易漏**。

lingyu9495-source/cn-pdf-report-typeset

中文报告做成 PDF,常见翻车三件套:**字体乱码、表格错位、字号小到客户在手机上根本看不清**。这套是交付级的排版标准加可直接跑的模板。

lingyu9495-source/company-law

公司出事,十有八九出在**治理和控制权**上——股权结构一开始搭歪,后面融资、上市、分家全是坑。

lingyu9495-source/excel-keep-format-edit

客户拿来一张老报表说「就在原表上改个数,格式一点都不要动」——用 openpyxl 重建一次,字体、边框、合并单元格、条件格式、列宽全丢,客户一打开就知道这不是他原来那张表。

lingyu9495-source/feasibility-report-engine

做可研报告,卡住的往往不是不会写,是**不知道按哪版大纲、财务测算六表和正文对不上、交付前排版被退稿**。这套是从二十多万字、131 张表的真实交付里固化的流水线。

lingyu9495-source/feasibility-study-consultant

项目可行性论证与研究报告写作框架 —— 任何「这事能不能做、怎么做、值不值得做」的商业决策场景:可行性研究/立项论证/新项目评估/创业方向验证。10大论证模块 + 强制联网尽调 + 多源交叉验证 + 读者视角写作 + 结构化报告输出。(v1.4 升级:从「引导客户把项目情况讲清楚」到「按问题类型与项目类型选论证方法」再到「用假设台账逐条验证」的六步教练流程;附可行性门槛评分器与假设台账两个零依赖脚本、项目卡与决策门清单模板)

lingyu9495-source/financial-statement-adjustment

存量财务报表在原表基础上改数字并保持三表勾稽关系。当用户要求「在原报表上改某些数字,其他数据跟着调整且格式完全不变」时使用:资产负债表/利润表/现金流量表的目标值调整、未分配利润与利润总额联动、勾稽校验。

lingyu9495-source/funding-fit-diagnosis

项目想融资,卡住的往往不是不会写 BP,而是不知道够不够格、该找谁、还缺什么、多久能到钱——问 AI 只得到"建议加强团队"这类正确的废话。这套引擎把它拆成可执行的四步:①10 问项目画像采集(缺关键项就先追问,不猜)②双轨评分:市场化 VC 适配分与政府基金适配分**分开算、各自判定**(≥75 强匹配 / 60-74 可争取 / 45-59 需补短板 / <45 暂不建议,低分不许只说"不行",要给替代路径)③点名机构:按赛道/阶段/地域从 23 家头部 VC 与产业资本(CVC)、21 家政府基金的卡片里逐条对照"匹配信号/劝退信号",每家写清"为什么是你、你还缺什么、材料怎么准备、递送路径",并单列"不建议碰"的机构省时间④差距清单(差距|补齐动作|耗时|难度|优先级,只挑 3 个月内能推动的)+ 三线找钱路线图(市场化 / 政府 / 替代路径)+ 时间预期与现金流提醒。包内另附 30 个真实融资案例对标、TS 条款红线、估值倍数参考、BP 制作铁律,以及零依赖离线评分脚本(纯标准库、不联网、不调大模型 API)。输入:项目情况(照着 10 问答即可);输出:一份研判报告(评分 + 优先接触名单 + 差距清单 + 路线图 + 风险提示),脚本侧可另出 JSON 评分结果。评分是适配度不是成功率,不承诺融资结果。触发词:融资、找投资、融资研判、能不能融到钱、找哪家风投、风险投资、VC、产业资本、CVC、政府引导基金、政府基金申报、产业基金、招商返投、项目申报、补贴申报、商业计划书、BP、路演、估值、股权稀释、TS 条款、对赌、回购、FA、财务顾问、找钱、融资渠道、天使轮、Pre-A、A轮。

lingyu9495-source/gov-fund-application

想拿政府的钱,最容易白跑半年——层级选错、赛道不在目录里、返投算不清,材料递上去才发现方向不对。这套参谋按可执行顺序走:①先判层级(国家级/省级/市级/区级,门槛与目的完全不同)②赛道对目录:不在目录里直接如实说没戏,不让用户白跑③21 家基金卡片逐家对照(投资方式、返投比例、让利规则、申报条件、申报通道、真实案例),点名"你这个项目适合找谁"④返投博弈:算清落地要求,给"不想迁注册地"的三条合规路径⑤材料与时间表:申报材料清单、评审周期(6-12 个月是常态)与现金流提醒。与主技能 `funding-fit-diagnosis` 配套(做整体融资研判时用主技能,VC 与政府双轨分开算)。输入:项目基本情况与注册地/落地意愿;输出:层级判定 + 可申报基金名单 + 返投方案 + 材料清单 + 时间预期。不承诺拿到资金,申报须如实提交材料;政策口径标注截止时间,细则请向当地主管部门核实。触发词:政府基金、政府引导基金、产业基金、政府补贴、专项资金、项目申报、补贴申报、申报条件、返投、招商落地、母基金、投资方式、让利、国家级基金、省级引导基金、市级基金、科技成果转化、专精特新、科技型企业、中小企业发展基金。

lingyu9495-source/hermes-profile-browser-isolation

多个 AI 助手同时干活时,浏览器会互相抢占:弹窗抢焦点、任务互相打断、你正在用的 Chrome/Edge 被顶掉——**你连自己的网页都点不动了**。

lingyu9495-source/investment-finance

做投融资最容易踩的坑不是"不懂概念",是**条款里的每一个字都在分钱**——估值方法选错、对赌触发条件写松、回购条款没兜底,签完才发现少拿几千万。

lingyu9495-source/lowvram-ai-video-comfyui

网上的 AI 视频教程开口就是 24G 显存,一看就把人劝退——**其实 8G 的 4060 也能跑**。这套是 RTX 4060 Laptop 8G 显存 / 32G 内存 / Win11 上真跑出来的配置与调参。

lingyu9495-source/novel-deai-detector

写完的稿子发布前想知道会不会被判 AI 生成、平台审核会不会卡、读者会不会一眼看出机器味——这套是检测到清洗的完整闭环。

lingyu9495-source/npl

不良资产这行的钱,赚在**买入价**上——买贵了,后面怎么处置都白搭。

lingyu9495-source/subagent-first

AI 助手不是不够聪明,是被重活占住了:你让它整理 40 个文件,它一头扎进去十几分钟不抬头——你想问进度、想补需求、想改方向,只能干等。这套作业纪律解决的就是这件事:凡是预计要跑多轮的活(多文件/长研究/批量/长搜索),一律派给后台子代理并行执行,主代理只保留四个动作——拆解、派活、汇总、对话,永远保留随时回你话的能力。我们自己在多智能体团队里天天跑,踩过的 12 个坑全部写进正文:子代理自报成功却发现产出是空的、context 没写全导致跑偏返工、两个子代理并行写同一文件互相覆盖、把需要用户拍板的活派给问不了人的子代理、跨会话长任务错用子代理导致某天悄悄停掉……① 决策树四问判定轻重(一轮工具调用能做完吗/需要用户拍板吗/要活过本次会话吗/能拆成互不依赖的块吗)② 派活四要素(goal / context 必须自包含——子代理看不到你和用户的对话 / output_schema / 验收方式)③ 五步流程:拆解→派活→并行调度→验收→收口 ④ 验收清单:存在性/合规性/真实性/完整性四组逐项打勾,并随机抽一处回一手来源核对 ⑤ 6 种委派模式 + 12 条反模式。配套 scripts/delegation_planner.py:纯标准库零依赖,喂一份任务清单(轮数/文件数/是否研究/是否批量/是否需拍板/是否不可逆/是否跨会话),直接输出「该派子代理 / 自己做 / 转定时任务」+ 并行分组建议 + 主代理必做清单。正文按平台无关的作业纪律写,并附常见平台工具名对照表;Hermes 的字段名/并发上限/后台语义另附一页实况对齐,其他平台按对照表映射即可。触发词:子代理、委派、delegate、任务编排、多智能体、并行执行、上下文管理、主代理被占住、AI不回话、等太久、插话卡住、批量处理、长研究、fork-join、fan-out。

Related Skills