altimate-data-engineering-skills

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Altimate Data Engineering Skills

Altimate 数据工程技能

Skill by ara.so — Data Skills collection.

Altimate Data Engineering Skills is a collection of Claude Code skills that encode the workflows and best practices of experienced analytics engineers. These skills transform Claude from a code generator into a capable data engineering assistant by teaching how to approach tasks, not just what syntax to use.

The project demonstrates 53% accuracy on ADE-bench (43 real-world dbt tasks), 3x improvement on model creation tasks, and 84% pass rate on Snowflake query optimization.

由ara.so开发的技能——数据技能合集。

Altimate 数据工程技能是一系列Claude Code技能，编码了资深分析工程师的工作流与最佳实践。这些技能通过教授如何处理任务（而非仅使用什么语法），将Claude从代码生成工具转变为得力的数据工程助手。

该项目在ADE-bench（43个真实dbt任务）上实现了53%的准确率，模型创建任务性能提升3倍，Snowflake查询优化通过率达84%。

What This Project Does

项目功能

Data Engineering Skills provides:

7 dbt skills: Model creation, debugging, testing, documentation, migration, refactoring, incremental models
3 Snowflake skills: Cost analysis, query optimization by ID, query optimization by text
1 delegation skill: Hand off complex tasks to altimate-code CLI tool
Workflow automation: Skills trigger automatically based on user intent
Best practices: Encoded patterns from experienced analytics engineers

Skills are markdown files with YAML frontmatter that define trigger conditions and step-by-step workflows.

数据工程技能提供：

7项dbt技能：模型创建、调试、测试、文档编写、迁移、重构、增量模型开发
3项Snowflake技能：成本分析、按ID优化查询、按文本优化查询
1项委托技能：将复杂任务交给altimate-code CLI工具处理
工作流自动化：根据用户意图自动触发技能
最佳实践：融入资深分析工程师的编码模式

技能是带有YAML前置元数据的markdown文件，用于定义触发条件和分步工作流。

Installation

安装方法

Installing Skills in Claude Code

在Claude Code中安装技能

bash

undefined

bash

undefined

Add the marketplace plugin

添加市场插件

/plugin marketplace add AltimateAI/data-engineering-skills

Install all skills

安装所有技能

/plugin install dbt-skills@data-engineering-skills /plugin install snowflake-skills@data-engineering-skills /plugin install altimate-code@data-engineering-skills

undefined

/plugin install dbt-skills@data-engineering-skills /plugin install snowflake-skills@data-engineering-skills /plugin install altimate-code@data-engineering-skills

undefined

Installing Kits

安装工具包

Kits bundle skills, MCP servers, and instructions:

bash

undefined

工具包整合了技能、MCP服务器和使用说明：

bash

undefined

Install altimate-code CLI (required for kits)

安装altimate-code CLI（工具包必备）

npm install -g altimate-code

Install kit system

安装工具包系统

altimate-code kit install AltimateAI/data-engineering-skills

Activate the dbt-snowflake kit

激活dbt-snowflake工具包

altimate-code kit activate dbt-snowflake

Check status

检查状态

altimate-code kit status

undefined

altimate-code kit status

undefined

Manual Installation

手动安装

Clone and reference skills directly:

bash

git clone https://github.com/AltimateAI/data-engineering-skills.git
cd data-engineering-skills

克隆仓库并直接引用技能：

bash

git clone https://github.com/AltimateAI/data-engineering-skills.git
cd data-engineering-skills

Available Skills

可用技能

dbt Skills

dbt技能

creating-dbt-models: Creates new dbt models following project conventions

Discovers existing patterns before writing
Runs
```
dbt build
```
after creation (not just compile)
Verifies output with
```
dbt show
```
Handles staging, intermediate, and mart models

debugging-dbt-errors: Troubleshoots dbt compilation and runtime errors

Reads full error messages carefully
Checks upstream dependencies
Applies fixes and rebuilds
Stops after 3 failed attempts to reassess

testing-dbt-models: Adds schema tests to models

Studies existing test patterns in project
Matches project testing style
Covers uniqueness, not_null, relationships, accepted_values

documenting-dbt-models: Generates dbt model documentation

Analyzes model logic and columns
Creates descriptions for models and fields
Follows project documentation patterns

migrating-sql-to-dbt: Converts legacy SQL to dbt models

Parses raw SQL queries
Creates proper dbt model structure
Handles CTEs, refs, and sources

refactoring-dbt-models: Safely restructures dbt models

Tracks all dependencies before changes
Applies refactoring
Verifies downstream models still work

developing-incremental-models: Creates incremental dbt models

Selects appropriate strategy (append, merge, delete+insert)
Designs proper unique_key
Handles late-arriving data and edge cases

creating-dbt-models：遵循项目规范创建新的dbt模型

编写前先发现现有模式
创建后运行
```
dbt build
```
（而非仅compile）
使用
```
dbt show
```
验证输出
处理staging、intermediate和mart模型

debugging-dbt-errors：排查dbt编译和运行时错误

仔细阅读完整错误信息
检查上游依赖
应用修复并重新构建
失败3次后停止操作以重新评估

testing-dbt-models：为模型添加schema测试

研究项目中已有的测试模式
匹配项目测试风格
覆盖唯一性、非空、关联关系、可接受值等测试

documenting-dbt-models：生成dbt模型文档

分析模型逻辑和列
为模型和字段创建描述
遵循项目文档规范

migrating-sql-to-dbt：将传统SQL转换为dbt模型

解析原始SQL查询
创建规范的dbt模型结构
处理CTE、refs和sources

refactoring-dbt-models：安全重构dbt模型

变更前追踪所有依赖
应用重构操作
验证下游模型仍可正常工作

developing-incremental-models：创建增量dbt模型

选择合适的策略（append、merge、delete+insert）
设计合理的unique_key
处理延迟到达的数据和边缘情况

Snowflake Skills

Snowflake技能

finding-expensive-queries: Identifies costly Snowflake queries

Finds queries by cost, time, or data scanned
Ranks by impact
Provides query IDs for optimization

optimizing-query-by-id: Optimizes using Snowflake query history ID

Retrieves query profile from ID
Applies optimization patterns
Validates semantic preservation

optimizing-query-text: Optimizes raw SQL query text

Profiles query execution
Identifies bottlenecks (scans, joins, aggregations)
Applies anti-pattern fixes
Tests performance improvement

finding-expensive-queries：识别高成本Snowflake查询

按成本、时间或扫描数据量查找查询
按影响程度排序
提供用于优化的查询ID

optimizing-query-by-id：通过Snowflake查询历史ID优化查询

根据ID获取查询配置文件
应用优化模式
验证语义一致性

optimizing-query-text：优化原始SQL查询文本

分析查询执行情况
识别瓶颈（扫描、连接、聚合）
修复反模式
测试性能提升效果

Delegation Skill

委托技能

altimate-code: Delegates complex data tasks to altimate-code CLI

Verifies altimate-code installation
Invokes
```
altimate-code run --yolo
```
non-interactively
Reads output file and summarizes results
Requires:
```
npm install -g altimate-code
```
(Node 20+)

altimate-code：将复杂数据任务委托给altimate-code CLI

验证altimate-code是否已安装
以非交互方式调用
```
altimate-code run --yolo
```
读取输出文件并总结结果
要求：
```
npm install -g altimate-code
```
（Node 20+）

Skill Structure

技能结构

Skills are markdown files with YAML frontmatter:

yaml

---
name: creating-dbt-models
description: |
  Guide for creating dbt models. ALWAYS use this skill when:
  (1) Creating ANY new model (staging, intermediate, mart)
  (2) Task mentions "create", "build", "add" with model/table
  (3) Modifying model logic or columns
---

Followed by workflow instructions:

markdown

undefined

技能是带有YAML前置元数据的markdown文件：

yaml

---
name: creating-dbt-models
description: |
  创建dbt模型的指南。在以下情况必须使用此技能：
  (1) 创建任何新模型（staging、intermediate、mart）
  (2) 任务提及与模型/表相关的"create"、"build"、"add"
  (3) 修改模型逻辑或列
---

后续是工作流说明：

markdown

undefined

dbt Model Development

dbt模型开发

Read before you write. Build after you write. Verify your output.

先阅读再编写，编写后构建，验证输出结果。

Critical Rules

关键规则

ALWAYS run
```
dbt build
```
after creating models - compile is NOT enough
ALWAYS verify output after build using
```
dbt show
```
If build fails 3+ times, stop and reassess your approach

创建模型后必须运行
```
dbt build
```
——仅compile不够
构建后必须使用
```
dbt show
```
验证输出
如果构建失败3次以上，停止操作并重新评估方法

Workflow

工作流

Discover Conventions
- Check existing models in same layer (staging/intermediate/mart)
- Note naming patterns, CTE style, column ordering
Write Model
- Follow discovered patterns
- Use proper refs and sources

Build and Verify

bash

dbt build --select model_name
dbt show --select model_name --limit 10

undefined

发现规范
- 检查同层级（staging/intermediate/mart）的现有模型
- 记录命名模式、CTE风格、列顺序
编写模型
- 遵循已发现的模式
- 使用正确的refs和sources

构建并验证

bash

dbt build --select model_name
dbt show --select model_name --limit 10

undefined

Usage Examples

使用示例

Creating a dbt Model

创建dbt模型

yaml

undefined

yaml

undefined

User request:

用户请求:

"Create a staging model for raw customers data"

"为原始客户数据创建staging模型"

Skill triggers: creating-dbt-models

触发技能: creating-dbt-models

AI workflow:

AI工作流:

1. Checks models/staging/ for naming patterns

1. 检查models/staging/目录下的命名模式

2. Creates models/staging/stg_customers.sql

2. 创建models/staging/stg_customers.sql

3. Runs: dbt build --select stg_customers

3. 运行: dbt build --select stg_customers

4. Verifies: dbt show --select stg_customers

4. 验证: dbt show --select stg_customers


Example model created:

```sql
-- models/staging/stg_customers.sql
with source as (
    select * from {{ source('jaffle_shop', 'customers') }}
),

renamed as (
    select
        id as customer_id,
        first_name,
        last_name,
        email,
        created_at
    from source
)

select * from renamed


创建的示例模型：

```sql
-- models/staging/stg_customers.sql
with source as (
    select * from {{ source('jaffle_shop', 'customers') }}
),

renamed as (
    select
        id as customer_id,
        first_name,
        last_name,
        email,
        created_at
    from source
)

select * from renamed

Debugging dbt Errors

调试dbt错误

yaml

undefined

yaml

undefined

User request:

用户请求:

"Fix this error: Compilation Error in model customers (models/customers.sql)"

"修复这个错误：Compilation Error in model customers (models/customers.sql)"

Skill triggers: debugging-dbt-errors

触发技能: debugging-dbt-errors

AI workflow:

AI工作流:

1. Reads full error message

1. 读取完整错误信息

2. Checks upstream model dependencies

2. 检查上游模型依赖

3. Identifies issue (e.g., missing ref)

3. 识别问题（例如：缺失ref）

4. Applies fix

4. 应用修复

5. Rebuilds: dbt build --select customers

5. 重新构建: dbt build --select customers

undefined

undefined

Optimizing Snowflake Query

优化Snowflake查询

yaml

undefined

yaml

undefined

User request:

用户请求:

"This query is taking 5 minutes, can you optimize it?"

"这个查询耗时5分钟，能优化吗？"

Skill triggers: optimizing-query-text

触发技能: optimizing-query-text

AI workflow:

AI工作流:

1. Profiles query (EXPLAIN, execution stats)

1. 分析查询（EXPLAIN、执行统计）

2. Identifies bottlenecks (table scans, inefficient joins)

2. 识别瓶颈（表扫描、低效连接）

3. Applies optimizations (clustering, filters, join order)

3. 应用优化（聚类、过滤、连接顺序）

4. Tests and measures improvement

4. 测试并衡量性能提升


Example optimization:

```sql
-- Before (slow)
SELECT 
    o.order_id,
    c.customer_name,
    SUM(oi.amount) as total
FROM orders o
JOIN customers c ON o.customer_id = c.customer_id
JOIN order_items oi ON o.order_id = oi.order_id
WHERE o.order_date >= '2024-01-01'
GROUP BY 1, 2;

-- After (optimized)
SELECT 
    o.order_id,
    c.customer_name,
    SUM(oi.amount) as total
FROM orders o
INNER JOIN order_items oi ON o.order_id = oi.order_id
LEFT JOIN customers c ON o.customer_id = c.customer_id
WHERE o.order_date >= '2024-01-01'
  AND o.order_date < '2024-12-31'  -- Added upper bound for partition pruning
GROUP BY 1, 2;
-- Assumes orders table is clustered by order_date


优化示例：

```sql
-- 优化前（慢）
SELECT 
    o.order_id,
    c.customer_name,
    SUM(oi.amount) as total
FROM orders o
JOIN customers c ON o.customer_id = c.customer_id
JOIN order_items oi ON o.order_id = oi.order_id
WHERE o.order_date >= '2024-01-01'
GROUP BY 1, 2;

-- 优化后（快）
SELECT 
    o.order_id,
    c.customer_name,
    SUM(oi.amount) as total
FROM orders o
INNER JOIN order_items oi ON o.order_id = oi.order_id
LEFT JOIN customers c ON o.customer_id = c.customer_id
WHERE o.order_date >= '2024-01-01'
  AND o.order_date < '2024-12-31'  -- 添加上限以实现分区裁剪
GROUP BY 1, 2;
-- 假设orders表已按order_date聚类

Using Kits

使用工具包

bash

undefined

bash

undefined

Activate dbt-snowflake kit for a project

为项目激活dbt-snowflake工具包

cd /path/to/dbt-project altimate-code kit activate dbt-snowflake

This configures:

此操作会配置:

- All dbt and Snowflake skills

- 所有dbt和Snowflake技能

- MCP server for dbt (live project access)

- 用于dbt的MCP服务器（实时项目访问）

- Project-specific instructions

- 项目专属说明

Deactivate when done

使用完成后停用

altimate-code kit deactivate


Kit configuration (`.altimate/kits/active.yaml`):

```yaml
kit: dbt-snowflake
version: 1.0.0
skills:
  - creating-dbt-models
  - debugging-dbt-errors
  - testing-dbt-models
  - documenting-dbt-models
  - migrating-sql-to-dbt
  - refactoring-dbt-models
  - developing-incremental-models
  - finding-expensive-queries
  - optimizing-query-by-id
  - optimizing-query-text
mcp_servers:
  - dbt
instructions: |
  You are working on a dbt + Snowflake project.
  Always check project conventions before creating models.
  Run dbt build (not compile) after creating models.

altimate-code kit deactivate


工具包配置文件（`.altimate/kits/active.yaml`）：

```yaml
kit: dbt-snowflake
version: 1.0.0
skills:
  - creating-dbt-models
  - debugging-dbt-errors
  - testing-dbt-models
  - documenting-dbt-models
  - migrating-sql-to-dbt
  - refactoring-dbt-models
  - developing-incremental-models
  - finding-expensive-queries
  - optimizing-query-by-id
  - optimizing-query-text
mcp_servers:
  - dbt
instructions: |
  你正在处理dbt + Snowflake项目。
  创建模型前务必检查项目规范。
  创建模型后运行dbt build（而非compile）。

Configuration

配置方法

Skill Configuration

技能配置

Skills are auto-triggered based on user requests. No explicit configuration needed.

技能会根据用户请求自动触发，无需显式配置。

MCP Integration

MCP集成

Skills work best with Altimate MCP server for live project access:

json

// claude_desktop_config.json or similar
{
  "mcpServers": {
    "altimate-dbt": {
      "command": "npx",
      "args": ["-y", "@altimate/mcp-server-dbt"],
      "env": {
        "DBT_PROJECT_DIR": "/path/to/dbt/project",
        "DBT_PROFILES_DIR": "/path/to/.dbt"
      }
    }
  }
}

MCP tools available to skills:

```
dbt_project_info
```
: Project structure, model list, sources
```
dbt_model_details
```
: Column types, dependencies, compiled SQL
```
dbt_compile
```
: Compile models without CLI
```
snowflake_query_history
```
: Recent query executions and stats
```
snowflake_table_stats
```
: Row counts, clustering info

配合Altimate MCP服务器使用技能可获得最佳效果，实现实时项目访问：

json

// claude_desktop_config.json或类似文件
{
  "mcpServers": {
    "altimate-dbt": {
      "command": "npx",
      "args": ["-y", "@altimate/mcp-server-dbt"],
      "env": {
        "DBT_PROJECT_DIR": "/path/to/dbt/project",
        "DBT_PROFILES_DIR": "/path/to/.dbt"
      }
    }
  }
}

技能可用的MCP工具：

```
dbt_project_info
```
：项目结构、模型列表、数据源
```
dbt_model_details
```
：列类型、依赖关系、编译后的SQL
```
dbt_compile
```
：无需CLI即可编译模型
```
snowflake_query_history
```
：近期查询执行记录和统计数据
```
snowflake_table_stats
```
：行数、聚类信息

Common Patterns

常见模式

Pattern 1: Multi-Step Model Creation

模式1：多步骤模型创建

yaml

undefined

yaml

undefined

User: "Create a mart model for monthly revenue by customer"

用户："创建一个按客户统计月度收入的mart模型"

Skills auto-chains:

自动链式触发技能:

1. creating-dbt-models: Creates initial model

1. creating-dbt-models：创建初始模型

2. testing-dbt-models: Adds schema tests

2. testing-dbt-models：添加schema测试

3. documenting-dbt-models: Adds documentation

3. documenting-dbt-models：添加文档

undefined

undefined

Pattern 2: Debug-Fix-Verify Loop

模式2：调试-修复-验证循环

yaml

undefined

yaml

undefined

User: "My model won't compile"

用户："我的模型无法编译"

Skill: debugging-dbt-errors

触发技能: debugging-dbt-errors

Workflow:

工作流:

1. Read error

1. 读取错误信息

2. Check dependencies

2. 检查依赖关系

3. Apply fix

3. 应用修复

4. Rebuild

4. 重新构建

5. If still fails, repeat up to 3x

5. 如果仍失败，重复最多3次

6. If 3x fails, stop and reassess approach

6. 3次失败后，停止操作并重新评估方法

undefined

undefined

Pattern 3: Safe Refactoring

模式3：安全重构

yaml

undefined

yaml

undefined

User: "Break this large model into smaller ones"

用户："将这个大型模型拆分成多个小模型"

Skill: refactoring-dbt-models

触发技能: refactoring-dbt-models

Workflow:

工作流:

1. Map all downstream dependencies

1. 映射所有下游依赖

2. Create new intermediate models

2. 创建新的中间模型

3. Update refs in downstream models

3. 更新下游模型中的refs

4. Run dbt build on entire DAG

4. 对整个DAG运行dbt build

5. Verify no breakage

5. 验证无故障

undefined

undefined

Pattern 4: Incremental Model Development

模式4：增量模型开发

sql

-- Pattern: merge strategy with unique_key
{{ config(
    materialized='incremental',
    unique_key='event_id',
    merge_update_columns=['status', 'updated_at']
) }}

select
    event_id,
    user_id,
    event_type,
    status,
    created_at,
    updated_at
from {{ source('events', 'raw_events') }}
{% if is_incremental() %}
    where updated_at > (select max(updated_at) from {{ this }})
{% endif %}

sql

-- 模式：带unique_key的merge策略
{{ config(
    materialized='incremental',
    unique_key='event_id',
    merge_update_columns=['status', 'updated_at']
) }}

select
    event_id,
    user_id,
    event_type,
    status,
    created_at,
    updated_at
from {{ source('events', 'raw_events') }}
{% if is_incremental() %}
    where updated_at > (select max(updated_at) from {{ this }})
{% endif %}

Pattern 5: Query Optimization Workflow

模式5：查询优化工作流

python

undefined

python

undefined

Skill: optimizing-query-text

触发技能: optimizing-query-text

Steps encoded:

编码步骤:

1. Profile query

1. 分析查询

2. Identify bottleneck type:

2. 识别瓶颈类型:

- Full table scan → Add filters/clustering

- 全表扫描 → 添加过滤/聚类

- Inefficient join → Reorder, change type

- 低效连接 → 调整顺序、更改类型

- Large aggregation → Pre-aggregate or partition

- 大型聚合 → 预聚合或分区

3. Apply pattern-based fix

3. 应用基于模式的修复

4. Validate semantics unchanged

4. 验证语义不变

5. Measure improvement

5. 衡量性能提升

undefined

undefined

Troubleshooting

故障排除

Skills Not Triggering

技能未触发

Problem: Skills don't activate for your request

Solution:

Be explicit: "Create a dbt model" not "make a model"
Check skill installation:
```
/plugin list
```
Review trigger phrases in skill YAML
Use exact trigger phrases from skill description

问题：技能未响应你的请求激活

解决方案：

表述明确：使用“创建dbt模型”而非“做一个模型”
检查技能安装：运行
```
/plugin list
```
查看技能YAML中的触发短语
使用技能描述中的精确触发短语

dbt Build Failures

dbt构建失败

Problem: Models fail to build after creation

Solution:

Skill will auto-retry up to 3 times
After 3 failures, skill stops to reassess
Check
```
dbt debug
```
for connection issues
Verify refs and sources exist
Check for SQL syntax errors

问题：模型创建后无法构建

解决方案:

技能会自动重试最多3次
3次失败后，技能会停止操作以重新评估
运行
```
dbt debug
```
检查连接问题
验证refs和sources存在
检查SQL语法错误

MCP Server Not Connected

MCP服务器未连接

Problem: Skills can't access live project data

Solution:

bash

undefined

问题：技能无法访问实时项目数据

解决方案:

bash

undefined

Check MCP server config

检查MCP服务器配置

cat claude_desktop_config.json | grep altimate

Verify dbt project path

验证dbt项目路径

export DBT_PROJECT_DIR=/correct/path export DBT_PROFILES_DIR=/correct/.dbt/path

Restart Claude Code

重启Claude Code

undefined

undefined

altimate-code Delegation Fails

altimate-code委托失败

Problem: "altimate-code not found" error

Solution:

bash

undefined

问题：出现“altimate-code not found”错误

解决方案:

bash

undefined

Install altimate-code CLI

安装altimate-code CLI

npm install -g altimate-code

Verify installation

验证安装

altimate-code --version

Check Node version (requires 20+)

检查Node版本（要求20+）

node --version

undefined

node --version

undefined

Query Optimization No Improvement

查询优化无提升

Problem: Optimized query performs the same

Solution:

Skill checks for anti-patterns but can't fix all issues
May need manual clustering/indexing setup
Check if bottleneck is data volume (consider aggregation)
Review Snowflake warehouse size
Use
```
finding-expensive-queries
```
to compare before/after

问题：优化后的查询性能无变化

解决方案:

技能会检查反模式，但无法解决所有问题
可能需要手动设置聚类/索引
检查瓶颈是否为数据量过大（考虑聚合）
检查Snowflake仓库大小
使用
```
finding-expensive-queries
```
对比优化前后

Kit Activation Issues

工具包激活失败

Problem: Kit won't activate

Solution:

bash

undefined

问题：工具包无法激活

解决方案:

bash

undefined

Check kit is installed

检查工具包是否已安装

altimate-code kit list

Reinstall if needed

必要时重新安装

altimate-code kit install AltimateAI/data-engineering-skills

Check for conflicting active kits

检查是否存在冲突的激活工具包

altimate-code kit status

Deactivate others first

先停用其他工具包

altimate-code kit deactivate

undefined

altimate-code kit deactivate

undefined

Performance

性能表现

Benchmark results (ADE-bench, 43 real-world dbt tasks):

Baseline Claude: 46.5% accuracy (20/43 tasks)
Claude + Skills: 53.5% accuracy (23/43 tasks)
Model Creation: 3x improvement (40% → 65%)

Snowflake optimization (TPC-H SF1000, 62 queries):

Baseline: 77.4% pass rate, 4.7% avg improvement
With Skills: 83.9% pass rate, 16.8% avg improvement (3.6x better)

基准测试结果（ADE-bench，43个真实dbt任务）：

基础Claude：46.5%准确率（20/43任务）
Claude + 技能：53.5%准确率（23/43任务）
模型创建：性能提升3倍（40% → 65%）

Snowflake优化测试（TPC-H SF1000，62个查询）：

基础版本：77.4%通过率，平均提升4.7%
搭配技能：83.9%通过率，平均提升16.8%（性能提升3.6倍）

Additional Resources

额外资源

Documentation: https://docs.myaltimate.com/
ADE-bench Framework: https://github.com/dbt-labs/ade-bench
altimate-code CLI: https://github.com/AltimateAI/altimate-code
dbt Slack: https://getdbt.slack.com/archives/C05KPDGRMDW
Contact: https://app.myaltimate.com/contactus

文档：https://docs.myaltimate.com/
ADE-bench框架：https://github.com/dbt-labs/ade-bench
altimate-code CLI：https://github.com/AltimateAI/altimate-code
dbt Slack：https://getdbt.slack.com/archives/C05KPDGRMDW
联系方式：https://app.myaltimate.com/contactus