Aivc

面向单细胞和 bulk 转录组的虚拟基因敲除、虚拟基因过表达与虚拟药物筛选流程 skill。当用户希望用已有数据运行虚拟干预、预测药物转录响应、生成逐基因/网络/embedding 结果、绘图或整理运行产物时使用。开始新分析时必须先请用户指定分析数据、分析目录、运行环境、KO/OE/药筛任务、模型及目标基因或药物库，缺项先问；Skill 完成输入检查、后端执行、结果标准化和文件交付。

7 stars 1 forksAI, LLMs & Agents

Install

skills CLI (Claude Code, Cursor, Codex and more)

npx skills add Pengzhi-Gao/aivc-skill --skill aivc

Manual install

git clone --depth 1 https://github.com/Pengzhi-Gao/aivc-skill.git /tmp/aivc-skill
mkdir -p ~/.claude/skills
cp -r /tmp/aivc-skill/. ~/.claude/skills/aivc

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# AIVC：虚拟干预流程

## 目标

将用户数据转换为可复核的虚拟干预结果。执行用户选择的模型，交付原始输出、标准化表格、图、参数和运行状态。

Skill 包含触发/调用规则、完整分析流程和示例代码。模型资源由执行者按规则在运行时获取和配置，复用用户已有的兼容资源；具体入口见 [模型资源与执行入口](references/model-setup.md)。

## 触发任务

- 虚拟基因敲除（KO）或基因调控网络扰动
- 虚拟基因过表达（OE）或 CRISPRa 条件响应
- 虚拟药物筛选、药物转录响应预测、疾病表达变化比较
- 对上述结果进行通路富集、状态/轨迹可视化、剂量比较或结果整理

## 强制工作流

### 1. 必须先完成启动问答

**每次新分析的第一次回复，先核对以下六项，把尚未明确的项合并提问；六项及所选模型的必要输入齐全后才能运行。** 沿用用户在当前任务中已明确指定的内容。现有文件、已安装模型和历史示例作为候选供用户选择。用户只说“运行一下”或“测试一下”时也执行本规则。修改 Skill、解释方法或查看已有结果时直接处理相应请求。

使用以下模板，只保留缺项，并列出少量可选文件的完整路径、格式及已知用途供用户选择：

```text
请补充本次分析设置（已给出的内容会沿用）：
1. 分析数据：文件完整路径；多个文件请说明基线、训练数据或疾病参照等用途。
2. 分析目录：本次工作文件和结果保存到哪个目录？
3. 运行环境：使用哪个现有环境/解释器？如需新建，请指定安装位置。
4. 分析任务：虚拟敲除（KO）/ 虚拟过表达（OE）/ 虚拟药物筛选。
5. 使用模型：KO/OE 可选 CellOracle 或 Geneformer；药筛按数据列出兼容后端供选择。
6. 干预对象：KO/OE 填目标基因（一个或列表）；药筛填药物库文件路径或候选药物清单。
```

任务已知时，第 5、6 项只展示该任务的选项。KO/OE 至少说明两个独立选项：CellOracle 输出 GRN 模拟表达变化与状态位移；Geneformer 输出排序编码扰动后的 embedding 影响。只有具备匹配的扰动训练数据时才列 GEARS。药筛按 `references/model-capabilities.md` 列出兼容的 PRnet、chemCPA、UniPert-G2CP、UniCure 或 DrugReflector，说明基线来源和输出类型；兼容性待核对的选项明确标注，由用户选择。用户选择固定背景模型时，数据项改为明确选定模型自带基线及细胞系；需要疾病比较时另选疾病参照。

分别记录分析目录、数据目录和环境目录。用户给出环境根目录时，先检查其中与所选模型匹配的解释器；能唯一确定时记录解析结果，有多个可行环境时请用户选择。新建环境按用户指定的位置和已有安装授权执行。文档中的历史路径作为候选供用户选择。

六项齐全后，简要回显有效设置并检查输入；无需再索要一次“确认开始”。KO/OE 要区分单基因逐个运行与联合扰动，用户给出多个基因但未说明时补问。按模型补齐物种、基因 ID、bulk/scRNA 类型、细胞背景、GRN/权重及剂量等必要字段；可从文件核实的元数据直接读取并记录来源。CellOracle OE 还需用户指定目标值或分位数及依据，Geneformer OE 不要求数值表达倍数。仅做干预响应不强制疾病数据；需要恢复方向比较时必须有疾病参照或可核对的差异基因表。

Read the full skill on GitHub

Details

Skill name: aivc
Repository: Pengzhi-Gao/aivc-skill
License: MIT
Repo updated: 16 October 2026

FAQ

What does the Aivc skill do?

How do I install the Aivc skill?

skills CLI (Claude Code, Cursor, Codex and more): npx skills add Pengzhi-Gao/aivc-skill --skill aivc. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is Aivc popular?

It lives in Pengzhi-Gao/aivc-skill, which has 7 GitHub stars and 1 forks.

Related skills

Bottle

Simooo45/bottle

Bootstrap a new or existing project with a small, file-based documentation system so Claude Code keeps context, lessons, and decisions across session restarts instead of starting cold every time. Sets up a handful of markdown files under .claude/ (standing rules, a free-form note to your next session, a phased roadmap, a prioritized backlog, and searchable history/changelog archives) sized and split so the files read every session stay small while the archives grow without ever costing context. Use this whenever someone is starting a new project and asks how to structure it for long-running AI

AI, LLMs & Agents

Marinara Engine Expert

cha1latte/marinara-engine-expert-skill

Expert for Marinara Engine (the local AI roleplay/chat frontend at github.com/Pasta-Devs/Marinara-Engine). Handles two kinds of work — IDEATION (designing characters, lorebooks, custom tools, agents, extensions for users building things in Marinara) and CONTRIBUTION (triaging PRs, reproducing bugs, and shipping focused changes to the Marinara codebase itself). Use this skill when the user mentions Marinara Engine, Professor Mari, SpicyMarinara, or describes a project involving AI characters, chatbots, roleplay assistants, or personas they want to build in a Marinara-Engine-style system. Also t

AI, LLMs & Agents

Changshu Arnold

factnn/changshu-arnold-skill

当用户提到"常熟阿诺"、"诺神"、"诺言诺语"、"那我问你"、"三卡车"、"生米冲碳"、"申气"、"脑袋尖尖"、"无糖可乐"、健身抽象梗时自动触发。将 Claude 变成常熟阿诺风格，用诺言诺语系统回答。即使用户只是随口提到这些词，也应该触发此 skill。

AI, LLMs & Agents

Global Think Tank Analyst

vassiliylakhonin/global-think-tank-analyst

Strategic-risk analysis skill for AI agents producing decision-ready memos on geopolitics, sanctions, trade, regulation, and strategic risk with explicit evidence boundaries, uncertainty, scenarios, actor incentives, trade-offs, and watch-next indicators. Use for a country risk brief, policy memo, sanctions or export-control exposure assessment, trade or regulatory implications memo, geopolitical scenario brief, strategic implications note for leadership, stakeholder and incentives analysis, a red-team challenge to an existing policy or risk view, or a decision briefing pack for founders, oper

AI, LLMs & Agents

Genpark Autonomous Personal Delegator Skill

Alpha-Park/genpark-autonomous-personal-delegator-skill

GenPark AI Agent Skill - Autonomous micro-action delegation with user guardrails, confidence thresholding, dry-run previews, and token budget gating.

AI, LLMs & Agents

Charming

tambo-labs/charming-mcp

Create, update, inspect, or troubleshoot Charming-hosted interactive web apps via the Charming MCP server, or connect an AI client (Claude, Cursor, Codex, Gemini CLI, ChatGPT) to Charming.

AI, LLMs & Agents

Browse the most popular Claude skills, all ai, llms & agents skills, or write your own with the skills generator.