airweave-search

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Airweave Search

Airweave搜索

Use this skill to effectively search and retrieve context from Airweave collections, whether answering questions or gathering context to complete tasks.

使用本技能可高效地从Airweave集合中搜索并检索上下文信息，无论是用于回答问题还是收集上下文以完成任务。

When to Search

何时进行搜索

Search when the user:

Asks about data in their connected apps ("What did we discuss in Slack about...")
Needs to find documents, messages, issues, or records
Asks factual questions about their workspace ("Who is responsible for...", "What's our policy on...")
References specific tools by name ("in Notion", "on GitHub", "in Jira")
Needs recent information you don't have in your training
Needs you to check app data for context to complete a task ("check our Notion docs", "look at the Jira ticket", "see what we decided in Slack")

Don't search when:

User asks general knowledge questions (use your training)
User is asking how to SET UP Airweave (use
```
airweave-setup
```
skill instead)
User already provided all needed context in the conversation
The question is about Airweave itself, not data within it

在以下场景下进行搜索：

用户询问其连接应用中的数据（例如“我们在Slack中讨论了关于……的内容？”）
用户需要查找文档、消息、问题或记录
用户询问关于工作区的事实性问题（例如“谁负责……？”、“我们的……政策是什么？”）
用户提及特定工具名称（例如“在Notion中”、“在GitHub上”、“在Jira中”）
用户需要你训练数据中没有的最新信息
用户需要你查看应用数据以获取上下文来完成任务（例如“查看我们的Notion文档”、“查看Jira工单”、“看看我们在Slack中做出的决定”）

请勿在以下场景下搜索：

用户询问通用知识类问题（使用你的训练数据回答）
用户询问如何设置Airweave（请改用
```
airweave-setup
```
技能）
用户已在对话中提供了所有所需上下文
问题是关于Airweave本身，而非其中的数据

Query Formulation

查询语句构建

Extract Key Concepts

提取核心概念

Turn user intent into effective search queries:

User Says	Search Query
"What did Sarah say about the launch?"	"Sarah product launch"
"Find the API documentation"	"API documentation"
"Any bugs reported this week?"	"bug report issues"
"What's our refund policy?"	"refund policy customer"

将用户意图转化为有效的搜索查询语句：

用户提问	搜索查询语句
“Sarah关于发布说了什么？”	“Sarah product launch”
“找到API文档”	“API documentation”
“这周有报告任何bug吗？”	“bug report issues”
“我们的退款政策是什么？”	“refund policy customer”

Query Tips

查询技巧

Use natural language - Airweave uses semantic search, not keyword matching
Include context - "pricing feedback" is better than just "pricing"
Be specific but not too narrow - Start moderately specific, broaden if no results
Avoid filler words - Skip "please find", "can you search for"

使用自然语言 - Airweave采用语义搜索，而非关键词匹配
包含上下文 - “pricing feedback”比仅用“pricing”效果更好
具体但不过于局限 - 从适度具体的查询开始，如果没有结果再扩大范围
避免冗余词汇 - 跳过“please find”、“can you search for”这类词汇

Parameter Selection

参数选择

Choose parameters based on user intent:

User Intent	Parameters
Recent updates/conversations	`recency_bias: 0.7-0.9`
Finding a specific document	`search_method: "keyword"` or `"hybrid"`
General topic exploration	`search_method: "hybrid"` , higher `limit`
High-quality results only	`enable_reranking: true`
Quick direct answer	`response_type: "completion"`
Browse/see all matches	`response_type: "raw"` , `limit: 20-50`

根据用户意图选择参数：

用户意图	参数设置
最新更新/对话	`recency_bias: 0.7-0.9`
查找特定文档	`search_method: "keyword"` 或 `"hybrid"`
主题探索	`search_method: "hybrid"` ，设置更高的 `limit` 值
仅获取高质量结果	`enable_reranking: true`
快速获取直接答案	`response_type: "completion"`
浏览所有匹配结果	`response_type: "raw"` , `limit: 20-50`

Parameter Quick Reference

参数速查

Parameter	Values	When to Use
`recency_bias`	0-1	Higher = favor recent. Use 0.7+ for "recent", "latest", "this week"
`search_method`	hybrid/neural/keyword	`keyword` for exact terms, `neural` for concepts, `hybrid` for both
`response_type`	raw/completion	`completion` for direct answers, `raw` to show sources
`limit`	1-1000	Lower (5-10) for quick answers, higher (20-50) for exploration
`enable_reranking`	boolean	`true` for better relevance (slightly slower)
`expansion_strategy`	auto/llm/no_expansion	`auto` for most cases, `no_expansion` for exact queries

See PARAMETERS.md for detailed guidance.

参数	可选值	使用场景
`recency_bias`	0-1	值越高越偏向最新内容。当用户提及“recent”、“latest”、“this week”时，设置为0.7+
`search_method`	hybrid/neural/keyword	`keyword` 用于精确术语， `neural` 用于概念搜索， `hybrid` 结合两者
`response_type`	raw/completion	`completion` 用于直接获取答案， `raw` 用于查看来源
`limit`	1-1000	较小值（5-10）用于快速获取答案，较大值（20-50）用于主题探索
`enable_reranking`	boolean	`true` 用于获取更相关的结果（速度略慢）
`expansion_strategy`	auto/llm/no_expansion	`auto` 适用于大多数场景， `no_expansion` 用于精确查询

详细指南请查看PARAMETERS.md。

Handling Results

结果处理

Interpreting Scores

分数解读

Score	Meaning	Action
0.85+	Highly relevant	Use confidently
0.70-0.85	Likely relevant	Use with context
0.50-0.70	Possibly relevant	Mention uncertainty
Below 0.50	Weak match	Consider rephrasing query

分数	含义	操作建议
0.85+	高度相关	可放心使用
0.70-0.85	可能相关	结合上下文使用
0.50-0.70	潜在相关	提及结果存在不确定性
低于0.50	匹配度低	考虑重新表述查询语句

Synthesizing Answers

答案整合

When presenting results to users:

Lead with the answer - Don't start with "I found 5 results"
Cite sources - Mention where info came from ("According to your Slack conversation...")
Synthesize, don't dump - Combine relevant parts into coherent response
Acknowledge gaps - If results don't fully answer, say so

向用户展示结果时：

先给出答案 - 不要以“我找到了5个结果”开头
注明来源 - 提及信息的来源（例如“根据你的Slack对话……”）
整合信息，不要堆砌 - 将相关部分组合成连贯的回答
说明信息缺口 - 如果结果无法完全回答问题，请告知用户

Handling No/Poor Results

无结果/低质量结果处理

If search returns no results or low-quality matches:

Broaden the query - Remove specific terms, use more general concepts
Try different phrasing - Rephrase using synonyms or related terms
Increase limit - Fetch more results to find relevant matches
Check source availability - The data source might not be connected
Ask for clarification - User might have more context to share

如果搜索无结果或匹配结果质量较低：

扩大查询范围 - 删除特定术语，使用更通用的概念
尝试不同表述 - 使用同义词或相关术语重新表述
提高结果数量限制 - 获取更多结果以找到相关匹配项
检查数据源可用性 - 数据源可能未连接
请求用户澄清 - 用户可能有更多上下文可以提供

Finding the Search Tool

查找搜索工具

Airweave MCP tools follow the naming pattern

search-{collection-name}

. Look for tools matching this pattern in your available MCP tools.

Examples:

```
search-acmes-slack-k8v2x1
```
```
search-acmes-notion-p3m9q7
```
```
search-acmes-jira-w5n4r2
```

If no Airweave search tool is available:

The user may not have Airweave MCP configured
Ask if they have Airweave set up and connected to their AI assistant
Suggest using the
```
airweave-setup
```
skill for configuration help

Multiple collections: If multiple

search-*

tools are available, choose based on the collection name and the user's request. If unclear which to use, ask the user or try the most general-sounding one first.

Airweave MCP工具遵循

search-{collection-name}

的命名模式。请在可用的MCP工具中查找符合该模式的工具。

示例：

```
search-acmes-slack-k8v2x1
```
```
search-acmes-notion-p3m9q7
```
```
search-acmes-jira-w5n4r2
```

如果没有可用的Airweave搜索工具：

用户可能未配置Airweave MCP
询问用户是否已设置Airweave并将其连接到AI助手
建议使用
```
airweave-setup
```
技能进行配置

多个集合的情况： 如果有多个

search-*

工具可用，请根据集合名称和用户请求选择合适的工具。如果不确定使用哪个，可以询问用户或先尝试名称最通用的工具。

Calling the Search Tool

调用搜索工具

Use the

search-{collection}

MCP tool with your chosen parameters:

search-acmes-slack-k8v2x1({
  query: "customer feedback pricing",
  recency_bias: 0.7,
  limit: 10
})

search-acmes-notion-p3m9q7({
  query: "API authentication docs",
  search_method: "hybrid",
  enable_reranking: true
})

search-acmes-jira-w5n4r2({
  query: "What is our refund policy?",
  response_type: "completion"
})

使用

search-{collection}

MCP工具并设置所选参数：

search-acmes-slack-k8v2x1({
  query: "customer feedback pricing",
  recency_bias: 0.7,
  limit: 10
})

search-acmes-notion-p3m9q7({
  query: "API authentication docs",
  search_method: "hybrid",
  enable_reranking: true
})

search-acmes-jira-w5n4r2({
  query: "What is our refund policy?",
  response_type: "completion"
})

Examples

示例

See EXAMPLES.md for complete conversation examples showing effective search patterns.

完整的对话示例请查看EXAMPLES.md，展示了有效的搜索模式。