@bgicli/bgicli 2.1.1 → 2.2.1

package/README.md

@@ -1,12 +1,13 @@
 # BGI CLI
 
-**BGI CLI** 是一个独立的生物信息学 AI 终端工具，专为中国研究者设计。
+**BGI CLI** 是面向中国生物学研究者的 AI 终端工具，开箱即用，无需额外配置。
 
-- ✅ **独立工具** — 无需安装 OpenCode 或其他 AI 工具
-- ✅ **中国 AI 服务商** — DeepSeek、Kimi、通义千问、MiniMax
-- ✅ **21 个预装工作流** — 覆盖转录组、基因组、表观基因组、临床分析
-- ✅ **真正的工具调用** — 执行 bash 命令、读写文件、运行 R/Python 脚本
-- ✅ **单一命令** — 直接运行 `bgi`
+- ✅ **开箱即用** — `npm install -g @bgicli/bgicli` 即可
+- ✅ **内置 889 个技能** — 21 个生信工作流 + 868 个 OpenClaw 医学技能，自动安装
+- ✅ **智能技能路由** — 描述任务自动激活对应技能，无需手动搜索
+- ✅ **中国 AI 服务商** — 百炼(DashScope)聚合：Qwen3.5、DeepSeek、Kimi、MiniMax 等 20+ 模型
+- ✅ **真实工具调用** — 执行 bash、读写文件、运行 R/Python 脚本
+- ✅ **内网支持** — 可接入公司私有化部署的大模型
 
 ---
 
@@ -14,96 +15,178 @@
 
 ```bash
 # 需要 Node.js 18+
-npm install -g @bgicli/bgicli
+npm install -g @bgicli/bgicli --registry https://registry.npmjs.org
+```
 
-# 或从源码安装
-git clone https://github.com/zja2004/BGI-CLI.git
-cd BGI-CLI
-npm install
-npm run build
-npm link
+首次运行自动初始化工作流和技能库（约 16MB），无需额外操作。
+
+```bash
+bgi
 ```
 
+---
+
+## 卸载
+
+```bash
+# 卸载 npm 包
+npm uninstall -g @bgicli/bgicli
+
+# 删除本地数据（配置、工作流、技能库）
+# Linux / macOS
+rm -rf ~/.bgicli
+
+# Windows PowerShell
+Remove-Item -Recurse -Force "$env:USERPROFILE\.bgicli"
+```
+
+---
+
 ## 快速开始
 
 ```bash
-bgi
+bgi                    # 启动
+/connect               # 首次配置 API Key
+/cat                   # 浏览技能分类目录
+/sk deseq2             # 搜索并激活 DESeq2 工作流
+/help                  # 查看全部命令
 ```
 
-首次运行会提示选择 AI 服务商并输入 API Key。
+首次运行提示配置百炼 (DashScope) API Key：
+- 获取地址：[bailian.console.aliyun.com](https://bailian.console.aliyun.com/) → API Key 管理
 
 ---
 
 ## 支持的 AI 服务商
 
-| 服务商 | 命令 | 获取 API Key |
-|--------|------|-------------|
-| **DeepSeek** (默认) | `/provider deepseek` | [platform.deepseek.com](https://platform.deepseek.com/api_keys) |
-| **Kimi** | `/provider kimi` | [platform.moonshot.cn](https://platform.moonshot.cn/console/api-keys) |
-| **通义千问** | `/provider qwen` | [dashscope.aliyuncs.com](https://dashscope.aliyuncs.com) |
-| **MiniMax** | `/provider minimax` | [platform.minimax.chat](https://platform.minimax.chat) |
+### 百炼 · 阿里云 (DashScope) — 默认
+通过百炼统一接入多个国内主流模型：
+
+| 模型 | 命令 |
+|------|------|
+| Qwen3.5-plus（默认） | `/model qwen3.5-plus` |
+| Qwen3-235B | `/model qwen3-235b-a22b` |
+| DeepSeek-R1 | `/model deepseek-r1` |
+| DeepSeek-V3 | `/model deepseek-v3` |
+| Kimi-K2.5 | `/model kimi-k2.5` |
+| MiniMax-M2.5 | `/model MiniMax-M2.5` |
+| QwQ-Plus（推理） | `/model qwq-plus` |
+
+获取 API Key：[bailian.console.aliyun.com](https://bailian.console.aliyun.com/)
+
+### 内网私有化部署
+```bash
+/provider intranet     # 切换到内网 Qwen3-235B（无需 Key）
+```
+
+### 自定义 OpenAI 兼容服务
+```bash
+/connect custom        # 配置任意 vLLM / Ollama / FastChat 地址
+```
 
 ---
 
 ## 命令参考
 
+### 服务商 / 模型
 | 命令 | 说明 |
 |------|------|
-| `/provider <name>` | 切换 AI 服务商 |
+| `/provider <name>` | 切换服务商 (`bailian` / `intranet` / `custom`) |
 | `/model <name>` | 切换模型 |
-| `/models` | 列出当前服务商的可用模型 |
+| `/models` | 列出当前服务商所有可用模型 |
 | `/providers` | 列出所有服务商 |
 | `/connect [provider]` | 配置 API Key |
 | `/status` | 显示当前配置 |
-| `/clear` | 清空对话历史 |
-| `/help` | 显示帮助 |
-| `exit` / `quit` | 退出 |
 
----
-
-## 预装工作流 (21个)
-
-运行分析时，BGI CLI 自动读取对应的 SKILL.md 工作流指南：
+### 技能与工作流
+| 命令 | 说明 |
+|------|------|
+| `/cat` | 按领域浏览技能分类目录（11个领域） |
+| `/sk` | 列出全部技能（工作流 + OpenClaw Medical） |
+| `/sk <关键词>` | 搜索并激活技能（如 `/sk deseq2`、`/sk alphafold`） |
+| `/wf` | 同 `/sk`，别名 |
 
-### 转录组学
-- `bulk-rnaseq-counts-to-de-deseq2` — DESeq2 差异表达分析
-- `bulk-omics-clustering` — 样本/特征聚类
-- `scrnaseq-scanpy-core-analysis` — 单细胞分析 (Scanpy/Python)
-- `scrnaseq-seurat-core-analysis` — 单细胞分析 (Seurat/R)
-- `spatial-transcriptomics` — 空间转录组
-- `coexpression-network` — 共表达网络 (WGCNA)
-- `functional-enrichment-from-degs` — 功能富集 (GO/KEGG/GSEA)
-- `grn-pyscenic` — 基因调控网络 (pySCENIC)
+> **智能路由**：直接描述任务，BGI CLI 自动识别并激活对应技能。
 
-### 基因组学
-- `genetic-variant-annotation` — 变异注释
-- `gwas-to-function-twas` — GWAS → TWAS
-- `mendelian-randomization-twosamplemr` — 孟德尔随机化
-- `polygenic-risk-score-prs-catalog` — 多基因风险评分
-- `pooled-crispr-screens` — CRISPR 文库筛选
+### 对话管理
+| 命令 | 说明 |
+|------|------|
+| `/clear` | 清空对话历史（重置激活的技能） |
+| `/history` | 查看对话统计（轮次 / Token 估算） |
+| `/save [文件名]` | 保存对话为 Markdown 文件 |
+| `/think [on\|off]` | 切换思考模式（Qwen3 `/think` 前缀） |
 
-### 表观基因组
-- `chip-atlas-peak-enrichment / diff-analysis / target-genes` — ChIP-Atlas 分析
+### 文件与目录
+| 命令 | 说明 |
+|------|------|
+| `/cd <路径>` | 更改工作目录 |
+| `/cwd` | 显示当前工作目录 |
+| `/tools` | 列出 AI 可调用的工具 |
+| `@路径` | 消息中内嵌文件内容（如 `@data.csv 里有什么?`） |
 
-### 临床与流行病学
-- `clinicaltrials-landscape`, `literature-preclinical`
-- `experimental-design-statistics`, `lasso-biomarker-panel`, `pcr-primer-design`
+### 其他
+| 命令 | 说明 |
+|------|------|
+| `/help` | 显示帮助 |
+| `exit` / `quit` / `q` | 退出 |
 
 ---
 
-## 工作流安装
+## 内置技能库
+
+### 生物信息学工作流（21个）
+
+#### 转录组学
+| ID | 说明 |
+|----|------|
+| `bulk-rnaseq-counts-to-de-deseq2` | DESeq2 差异表达分析 |
+| `bulk-omics-clustering` | 样本/特征聚类（K-Means / HDBSCAN） |
+| `scrnaseq-scanpy-core-analysis` | 单细胞 RNA-seq（Scanpy/Python） |
+| `scrnaseq-seurat-core-analysis` | 单细胞 RNA-seq（Seurat/R） |
+| `spatial-transcriptomics` | 空间转录组（Visium） |
+| `coexpression-network` | 共表达网络（WGCNA） |
+| `functional-enrichment-from-degs` | 功能富集（GO / KEGG / GSEA） |
+| `grn-pyscenic` | 基因调控网络（pySCENIC） |
+
+#### 基因组学
+| ID | 说明 |
+|----|------|
+| `genetic-variant-annotation` | VCF 变异注释（VEP / ANNOVAR） |
+| `gwas-to-function-twas` | GWAS → TWAS 因果基因 |
+| `mendelian-randomization-twosamplemr` | 孟德尔随机化 |
+| `polygenic-risk-score-prs-catalog` | 多基因风险评分（PRS） |
+| `pooled-crispr-screens` | CRISPR 文库筛选（MAGeCK / BAGEL2） |
+
+#### 表观基因组
+| ID | 说明 |
+|----|------|
+| `chip-atlas-peak-enrichment` | ChIP-seq 峰值富集 |
+| `chip-atlas-diff-analysis` | 差异结合分析 |
+| `chip-atlas-target-genes` | 转录因子靶基因鉴定 |
+
+#### 临床与流行病学
+| ID | 说明 |
+|----|------|
+| `clinicaltrials-landscape` | 临床试验格局分析 |
+| `literature-preclinical` | 临床前文献系统提取 |
+| `experimental-design-statistics` | 实验设计与统计检验 |
+| `lasso-biomarker-panel` | LASSO 生物标志物筛选 |
+| `pcr-primer-design` | PCR/qPCR 引物设计 |
+
+### OpenClaw Medical Skills（868个）
+
+覆盖结构生物学、单细胞、药物发现、抗体设计、文献检索等领域，使用 `/cat` 浏览分类目录。
 
-将 21 个分析工作流部署到 `~/.bgicli/workflows/`：
-
-```bash
-# 从 bgicli-opencode 目录复制（如果已克隆旧仓库）
-cp -r /path/to/old/workflows ~/.bgicli/workflows/
-```
+---
 
-或运行安装脚本（Linux/macOS）：
+## 从源码安装
 
 ```bash
-bash install.sh
+git clone https://github.com/zja2004/BGI-CLI.git
+cd BGI-CLI
+npm install
+npm run build
+npm link
 ```
 
 ---
@@ -111,22 +194,17 @@ bash install.sh
 ## 架构
 
 ```
-bgi (单一二进制)
-├── src/index.ts     — CLI 主入口、命令处理、交互循环
-├── src/chat.ts      — 流式对话引擎（支持工具调用）
-├── src/tools.ts     — 工具实现 (bash, read_file, write_file, list_dir, search_files)
-├── src/prompt.ts    — 嵌入式生物信息学系统提示 + 工作流索引
-├── src/providers.ts — 中国 AI 服务商配置
-└── src/config.ts    — 配置管理 (~/.bgicli/config.json)
+bgi
+├── src/index.ts       — CLI 主入口、命令处理、智能路由
+├── src/chat.ts        — 流式对话引擎（工具调用循环）
+├── src/tools.ts       — 工具实现（bash / read_file / write_file 等）
+├── src/skillRouter.ts — 关键词路由表（35个核心技能自动匹配）
+├── src/prompt.ts      — 生物信息学系统提示
+├── src/providers.ts   — 中国 AI 服务商配置
+├── src/config.ts      — 配置管理（~/.bgicli/config.json）
+└── data/              — 内置数据（工作流 + Skills + Python 工具）
 ```
 
-**工具调用流程**：
-1. 用户提问 → 发给 LLM（带工具定义）
-2. LLM 决定调用工具（bash/read_file 等）
-3. BGI CLI 执行工具，将结果返回给 LLM
-4. LLM 基于执行结果继续回答
-5. 循环直到 LLM 完成回答
-
 ---
 
 ## License

package/data/skills/aav-vector-design-agent/SKILL.md

@@ -0,0 +1,198 @@
+<!--
+# COPYRIGHT NOTICE
+# This file is part of the "Universal Biomedical Skills" project.
+# Copyright (c) 2026 MD BABU MIA, PhD <md.babu.mia@mssm.edu>
+# All Rights Reserved.
+#
+# This code is proprietary and confidential.
+# Unauthorized copying of this file, via any medium is strictly prohibited.
+#
+# Provenance: Authenticated by MD BABU MIA
+
+-->
+
+---
+name: 'aav-vector-design-agent'
+description: 'AI-powered adeno-associated virus (AAV) vector design for gene therapy including capsid engineering, promoter selection, and tropism optimization.'
+measurable_outcome: Execute skill workflow successfully with valid output within 15 minutes.
+allowed-tools:
+  - read_file
+  - run_shell_command
+---
+
+
+# AAV Vector Design Agent
+
+The **AAV Vector Design Agent** provides AI-driven design of adeno-associated virus vectors for gene therapy applications. It covers capsid selection and engineering, promoter/enhancer design, transgene optimization, and manufacturing considerations.
+
+## When to Use This Skill
+
+* When selecting optimal AAV serotype for tissue-specific targeting.
+* To design novel capsid variants with enhanced properties.
+* For optimizing transgene expression cassettes.
+* When predicting immunogenicity and neutralizing antibody escape.
+* To design liver-detargeted or CNS-tropic vectors.
+
+## Core Capabilities
+
+1. **Capsid Selection**: Match AAV serotype to target tissue based on tropism profiles.
+
+2. **Capsid Engineering**: Design modified capsids for enhanced transduction or immune evasion.
+
+3. **Promoter Design**: Select and optimize tissue-specific or ubiquitous promoters.
+
+4. **Transgene Optimization**: Codon optimization and regulatory element design.
+
+5. **Immunogenicity Prediction**: Predict NAb binding and T-cell epitopes.
+
+6. **Manufacturing Assessment**: Evaluate producibility and purification considerations.
+
+## AAV Serotype Tropism
+
+| Serotype | Primary Tropism | Clinical Use |
+|----------|-----------------|--------------|
+| AAV1 | Muscle, CNS | Glybera (muscle) |
+| AAV2 | Broad (liver, muscle) | Luxturna (retina) |
+| AAV5 | CNS, liver, retina | Hemgenix (liver) |
+| AAV8 | Liver, muscle | Multiple trials |
+| AAV9 | CNS, cardiac, liver | Zolgensma (CNS) |
+| AAVrh10 | CNS, liver | CNS trials |
+| AAVrh74 | Muscle | Elevidys (muscle) |
+| AAV-PHP.eB | CNS (mouse) | Research |
+
+## Workflow
+
+1. **Input**: Target tissue, therapeutic gene, patient population characteristics.
+
+2. **Capsid Selection**: Rank serotypes by tropism profile match.
+
+3. **Capsid Engineering**: Design modifications if needed (peptide insertion, point mutations).
+
+4. **Cassette Design**: Optimize ITR-to-ITR expression cassette.
+
+5. **Immunogenicity Analysis**: Predict NAb prevalence and T-cell epitopes.
+
+6. **Manufacturing Review**: Assess production feasibility.
+
+7. **Output**: Complete vector design with rationale.
+
+## Example Usage
+
+**User**: "Design an AAV vector for liver-directed gene therapy in hemophilia B with low immunogenicity."
+
+**Agent Action**:
+```bash
+python3 Skills/Gene_Therapy/AAV_Vector_Design_Agent/aav_designer.py \
+    --target_tissue liver \
+    --therapeutic_gene F9 \
+    --indication hemophilia_b \
+    --minimize_immunogenicity true \
+    --nab_escape true \
+    --promoter liver_specific \
+    --output aav_design/
+```
+
+## Expression Cassette Components
+
+```
+5' ITR - [Promoter] - [5' UTR] - [Transgene] - [WPRE] - [PolyA] - 3' ITR
+
+Packaging limit: ~4.7 kb between ITRs
+```
+
+**Promoter Options**:
+| Promoter | Type | Size | Application |
+|----------|------|------|-------------|
+| CAG | Ubiquitous | 1.7 kb | Strong expression |
+| EF1α | Ubiquitous | 1.2 kb | Constitutive |
+| LP1 | Liver-specific | 0.5 kb | Hepatocyte targeting |
+| hSyn | Neuron-specific | 0.5 kb | CNS applications |
+| MCK | Muscle-specific | 0.6 kb | Myopathies |
+| CMV | Ubiquitous | 0.6 kb | High initial (silenced) |
+
+## Capsid Engineering Strategies
+
+**Directed Evolution**:
+- Error-prone PCR libraries
+- DNA shuffling
+- Selection in target tissue
+
+**Rational Design**:
+- Peptide display (insertion in variable loops)
+- Point mutations for receptor targeting
+- Tyrosine-to-phenylalanine for stability
+
+**Machine Learning**:
+- Sequence-function models
+- Generative models for novel capsids
+- Tropism prediction
+
+## Immunogenicity Considerations
+
+**Pre-existing NAbs**:
+| Serotype | NAb Prevalence |
+|----------|----------------|
+| AAV2 | 30-60% |
+| AAV5 | 15-30% |
+| AAV8 | 15-25% |
+| AAV9 | 20-35% |
+
+**Mitigation Strategies**:
+- Serotype selection based on patient screening
+- Engineered NAb-evading capsids
+- Immunosuppression protocols
+- Plasmapheresis
+
+## AI/ML Components
+
+**Tropism Prediction**:
+- CNN on capsid sequence
+- Cell-type specific transduction
+- Cross-species translation
+
+**Immunogenicity Modeling**:
+- MHC binding prediction
+- T-cell epitope mapping
+- NAb epitope prediction
+
+**Expression Optimization**:
+- Codon optimization algorithms
+- RNA structure prediction
+- miRNA target site avoidance
+
+## Manufacturing Considerations
+
+| Factor | Impact | Optimization |
+|--------|--------|--------------|
+| Capsid yield | Production cost | Sequence modifications |
+| Empty/full ratio | Potency | Purification method |
+| Aggregation | Stability | Formulation |
+| DNA packaging | Transgene size | Cassette design |
+
+## Prerequisites
+
+* Python 3.10+
+* Sequence analysis tools
+* Immunoinformatics packages
+* Structural biology tools
+
+## Related Skills
+
+* CRISPR_Design_Agent - For gene editing payloads
+* Protein_Engineering - For capsid design
+* RNA_Therapeutics - For alternative modalities
+
+## Regulatory Considerations
+
+1. **Biodistribution**: Required for IND
+2. **Shedding**: Vector in bodily fluids
+3. **Germline transmission**: Gonadal presence
+4. **Integration risk**: Random vs site-specific
+5. **Immunogenicity**: Pre-existing and induced
+
+## Author
+
+AI Group - Biomedical AI Platform
+
+
+<!-- AUTHOR_SIGNATURE: 9a7f3c2e-MD-BABU-MIA-2026-MSSM-SECURE -->

package/data/skills/adaptyv/SKILL.md

@@ -0,0 +1,112 @@
+---
+name: adaptyv
+description: Cloud laboratory platform for automated protein testing and validation. Use when designing proteins and needing experimental validation including binding assays, expression testing, thermostability measurements, enzyme activity assays, or protein sequence optimization. Also use for submitting experiments via API, tracking experiment status, downloading results, optimizing protein sequences for better expression using computational tools (NetSolP, SoluProt, SolubleMPNN, ESM), or managing protein design workflows with wet-lab validation.
+license: Unknown
+metadata:
+    skill-author: K-Dense Inc.
+---
+
+# Adaptyv
+
+Adaptyv is a cloud laboratory platform that provides automated protein testing and validation services. Submit protein sequences via API or web interface and receive experimental results in approximately 21 days.
+
+## Quick Start
+
+### Authentication Setup
+
+Adaptyv requires API authentication. Set up your credentials:
+
+1. Contact support@adaptyvbio.com to request API access (platform is in alpha/beta)
+2. Receive your API access token
+3. Set environment variable:
+
+```bash
+export ADAPTYV_API_KEY="your_api_key_here"
+```
+
+Or create a `.env` file:
+
+```
+ADAPTYV_API_KEY=your_api_key_here
+```
+
+### Installation
+
+Install the required package using uv:
+
+```bash
+uv pip install requests python-dotenv
+```
+
+### Basic Usage
+
+Submit protein sequences for testing:
+
+```python
+import os
+import requests
+from dotenv import load_dotenv
+
+load_dotenv()
+
+api_key = os.getenv("ADAPTYV_API_KEY")
+base_url = "https://kq5jp7qj7wdqklhsxmovkzn4l40obksv.lambda-url.eu-central-1.on.aws"
+
+headers = {
+    "Authorization": f"Bearer {api_key}",
+    "Content-Type": "application/json"
+}
+
+# Submit experiment
+response = requests.post(
+    f"{base_url}/experiments",
+    headers=headers,
+    json={
+        "sequences": ">protein1\nMKVLWALLGLLGAA...",
+        "experiment_type": "binding",
+        "webhook_url": "https://your-webhook.com/callback"
+    }
+)
+
+experiment_id = response.json()["experiment_id"]
+```
+
+## Available Experiment Types
+Adaptyv supports multiple assay types:
+- **Binding assays** - Test protein-target interactions using biolayer interferometry
+- **Expression testing** - Measure protein expression levels
+- **Thermostability** - Characterize protein thermal stability
+- **Enzyme activity** - Assess enzymatic function
+
+See `reference/experiments.md` for detailed information on each experiment type and workflows.
+
+## Protein Sequence Optimization
+Before submitting sequences, optimize them for better expression and stability:
+
+**Common issues to address:**
+- Unpaired cysteines that create unwanted disulfides
+- Excessive hydrophobic regions causing aggregation
+- Poor solubility predictions
+
+**Recommended tools:**
+- NetSolP / SoluProt - Initial solubility filtering
+- SolubleMPNN - Sequence redesign for improved solubility
+- ESM - Sequence likelihood scoring
+- ipTM - Interface stability assessment
+- pSAE - Hydrophobic exposure quantification
+
+See `reference/protein_optimization.md` for detailed optimization workflows and tool usage.
+
+## API Reference
+For complete API documentation including all endpoints, request/response formats, and authentication details, see `reference/api_reference.md`.
+
+## Examples
+For concrete code examples covering common use cases (experiment submission, status tracking, result retrieval, batch processing), see `reference/examples.md`.
+
+## Important Notes
+- Platform is currently in alpha/beta phase with features subject to change
+- Not all platform features are available via API yet
+- Results typically delivered in ~21 days
+- Contact support@adaptyvbio.com for access requests or questions
+- Suitable for high-throughput AI-driven protein design workflows
+