zotero-mcp-code

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Zotero MCP Code Execution Skill

Zotero MCP 代码执行技能

Search your Zotero library using code execution for safe, efficient, comprehensive searches.

通过代码执行安全、高效、全面地搜索你的Zotero库。

🎯 Core Concept

🎯 核心理念

Instead of calling MCP tools directly (which loads all results into context and risks crashes), write Python code that:

Fetches large datasets (50-100+ items per strategy)
Filters and ranks in code execution environment
Returns only top N results to context

Benefits:

✅ No crash risk (large data stays in code)
✅ Automatic multi-strategy search
✅ Automatic deduplication
✅ Automatic ranking
✅ One function call instead of 5-10

不同于直接调用MCP工具（会将所有结果加载到上下文环境中，存在崩溃风险），编写Python代码来完成以下操作：

获取大型数据集（每种策略获取50-100+条条目）
在代码执行环境中进行过滤和排序
仅将排名前N的结果返回至上下文环境

优势：

✅ 无崩溃风险（大型数据仅在代码中处理）
✅ 自动执行多策略搜索
✅ 自动去重
✅ 自动排序
✅ 一次函数调用替代5-10次调用

🚀 Basic Usage

🚀 基本用法

For 90% of Zotero searches, use this simple pattern:

python

import sys
sys.path.append('/Users/niyaro/Documents/Code/zotero-code-execution')
import setup_paths

from zotero_lib import SearchOrchestrator, format_results

对于90%的Zotero搜索场景，可使用以下简单模式：

python

import sys
sys.path.append('/Users/niyaro/Documents/Code/zotero-code-execution')
import setup_paths

from zotero_lib import SearchOrchestrator, format_results

Single comprehensive search

单次全面搜索

orchestrator = SearchOrchestrator() results = orchestrator.comprehensive_search( "user's query here", max_results=20 # Return top 20 most relevant )

orchestrator = SearchOrchestrator() results = orchestrator.comprehensive_search( "用户查询内容", max_results=20 # 返回相关性最高的20条结果 )

Format and display

格式化并展示结果

print(format_results(results, include_abstracts=True))


**This automatically:**
- Performs semantic search (multiple variations)
- Performs keyword search (multiple variations)
- Performs tag-based search
- Fetches 100+ items total
- Deduplicates results
- Ranks by relevance
- Returns only top 20 to context

print(format_results(results, include_abstracts=True))


**该代码会自动完成：**
- 执行语义搜索（包含多种变体）
- 执行关键词搜索（包含多种变体）
- 执行基于标签的搜索
- 总共获取100+条条目
- 对结果去重
- 按相关性排序
- 仅将前20条结果返回至上下文环境

📋 Common Patterns

📋 常见使用模式

Pattern 1: Simple Search (Most Common)

模式1：简单搜索（最常用）

User asks: "Find papers about embodied cognition"

python

import sys
sys.path.append('/Users/niyaro/Documents/Code/zotero-code-execution')
import setup_paths
from zotero_lib import SearchOrchestrator, format_results

orchestrator = SearchOrchestrator()
results = orchestrator.comprehensive_search("embodied cognition", max_results=20)
print(format_results(results))

用户需求： "查找关于具身认知的论文"

python

import sys
sys.path.append('/Users/niyaro/Documents/Code/zotero-code-execution')
import setup_paths
from zotero_lib import SearchOrchestrator, format_results

orchestrator = SearchOrchestrator()
results = orchestrator.comprehensive_search("embodied cognition", max_results=20)
print(format_results(results))

Pattern 2: Filtered Search

模式2：过滤搜索

User asks: "Find recent journal articles about machine learning"

python

import sys
sys.path.append('/Users/niyaro/Documents/Code/zotero-code-execution')
import setup_paths
from zotero_lib import ZoteroLibrary, SearchOrchestrator, format_results

library = ZoteroLibrary()
orchestrator = SearchOrchestrator(library)

用户需求： "查找近年关于机器学习的期刊文章"

python

import sys
sys.path.append('/Users/niyaro/Documents/Code/zotero-code-execution')
import setup_paths
from zotero_lib import ZoteroLibrary, SearchOrchestrator, format_results

library = ZoteroLibrary()
orchestrator = SearchOrchestrator(library)

Fetch broadly (safe - filtering happens in code)

广泛获取条目（安全操作 - 过滤在代码中完成）

items = library.search_items("machine learning", limit=100)

Filter in code

在代码中过滤

filtered = orchestrator.filter_by_criteria( items, item_types=["journalArticle"], date_range=(2020, 2025) )

print(format_results(filtered[:15]))

undefined

filtered = orchestrator.filter_by_criteria( items, item_types=["journalArticle"], date_range=(2020, 2025) )

print(format_results(filtered[:15]))

undefined

Pattern 3: Author Search

模式3：作者搜索

User asks: "What papers do I have by Kahneman?"

python

import sys
sys.path.append('/Users/niyaro/Documents/Code/zotero-code-execution')
import setup_paths
from zotero_lib import ZoteroLibrary, format_results

library = ZoteroLibrary()
results = library.search_items(
    "Kahneman",
    qmode="titleCreatorYear",
    limit=50
)

用户需求： "我库中有哪些Kahneman的论文？"

python

import sys
sys.path.append('/Users/niyaro/Documents/Code/zotero-code-execution')
import setup_paths
from zotero_lib import ZoteroLibrary, format_results

library = ZoteroLibrary()
results = library.search_items(
    "Kahneman",
    qmode="titleCreatorYear",
    limit=50
)

Sort by date

按日期排序

sorted_results = sorted(results, key=lambda x: x.date, reverse=True) print(format_results(sorted_results))

undefined

sorted_results = sorted(results, key=lambda x: x.date, reverse=True) print(format_results(sorted_results))

undefined

Pattern 4: Tag-Based Search

模式4：基于标签的搜索

User asks: "Show me papers tagged with 'learning' and 'cognition'"

python

import sys
sys.path.append('/Users/niyaro/Documents/Code/zotero-code-execution')
import setup_paths
from zotero_lib import ZoteroLibrary, format_results

library = ZoteroLibrary()
results = library.search_by_tag(["learning", "cognition"], limit=50)
print(format_results(results[:20]))

用户需求： "展示标记为'learning'和'cognition'的论文"

python

import sys
sys.path.append('/Users/niyaro/Documents/Code/zotero-code-execution')
import setup_paths
from zotero_lib import ZoteroLibrary, format_results

library = ZoteroLibrary()
results = library.search_by_tag(["learning", "cognition"], limit=50)
print(format_results(results[:20]))

Pattern 5: Recent Papers

模式5：最近添加的论文

User asks: "What did I recently add?"

python

import sys
sys.path.append('/Users/niyaro/Documents/Code/zotero-code-execution')
import setup_paths
from zotero_lib import ZoteroLibrary, format_results

library = ZoteroLibrary()
results = library.get_recent(limit=20)
print(format_results(results))

用户需求： "我最近添加了哪些内容？"

python

import sys
sys.path.append('/Users/niyaro/Documents/Code/zotero-code-execution')
import setup_paths
from zotero_lib import ZoteroLibrary, format_results

library = ZoteroLibrary()
results = library.get_recent(limit=20)
print(format_results(results))

Pattern 6: Multi-Topic Search

模式6：多主题搜索

User asks: "Find papers about both cognition and learning"

python

import sys
sys.path.append('/Users/niyaro/Documents/Code/zotero-code-execution')
import setup_paths
from zotero_lib import SearchOrchestrator, format_results

orchestrator = SearchOrchestrator()

用户需求： "查找同时关于认知和学习的论文"

python

import sys
sys.path.append('/Users/niyaro/Documents/Code/zotero-code-execution')
import setup_paths
from zotero_lib import SearchOrchestrator, format_results

orchestrator = SearchOrchestrator()

Search both topics

搜索两个主题

results1 = orchestrator.comprehensive_search("cognition", max_results=30) results2 = orchestrator.comprehensive_search("learning", max_results=30)

Find intersection

查找交集

keys1 = {item.key for item in results1} keys2 = {item.key for item in results2} common_keys = keys1 & keys2

if common_keys: common_items = [item for item in results1 if item.key in common_keys] print("Papers about both topics:") print(format_results(common_items)) else: print("No papers found on both topics.") print("\nCognition results:") print(format_results(results1[:10])) print("\nLearning results:") print(format_results(results2[:10]))

undefined

keys1 = {item.key for item in results1} keys2 = {item.key for item in results2} common_keys = keys1 & keys2

if common_keys: common_items = [item for item in results1 if item.key in common_keys] print("同时涉及两个主题的论文：") print(format_results(common_items)) else: print("未找到同时涉及两个主题的论文。") print("\n认知主题结果：") print(format_results(results1[:10])) print("\n学习主题结果：") print(format_results(results2[:10]))

undefined

🔧 Advanced Usage

🔧 高级用法

Custom Filtering Logic

自定义过滤逻辑

python

import sys
sys.path.append('/Users/niyaro/Documents/Code/zotero-code-execution')
import setup_paths
from zotero_lib import ZoteroLibrary, SearchOrchestrator, format_results

library = ZoteroLibrary()
orchestrator = SearchOrchestrator(library)

python

import sys
sys.path.append('/Users/niyaro/Documents/Code/zotero-code-execution')
import setup_paths
from zotero_lib import ZoteroLibrary, SearchOrchestrator, format_results

library = ZoteroLibrary()
orchestrator = SearchOrchestrator(library)

Fetch large dataset

获取大型数据集

items = library.search_items("neural networks", limit=100)

Custom filtering

自定义过滤

recent_with_doi = [ item for item in items if item.doi and item.date and int(item.date[:4]) >= 2020 ]

print(format_results(recent_with_doi[:15]))

undefined

recent_with_doi = [ item for item in items if item.doi and item.date and int(item.date[:4]) >= 2020 ]

print(format_results(recent_with_doi[:15]))

undefined

Multi-Angle Custom Search

多角度自定义搜索

python

import sys
sys.path.append('/Users/niyaro/Documents/Code/zotero-code-execution')
import setup_paths
from zotero_lib import ZoteroLibrary, SearchOrchestrator, format_results

library = ZoteroLibrary()
orchestrator = SearchOrchestrator(library)

all_results = set()

python

import sys
sys.path.append('/Users/niyaro/Documents/Code/zotero-code-execution')
import setup_paths
from zotero_lib import ZoteroLibrary, SearchOrchestrator, format_results

library = ZoteroLibrary()
orchestrator = SearchOrchestrator(library)

all_results = set()

Multiple search angles

多个搜索角度

queries = [ "skill transfer", "transfer of learning", "generalization of skills" ]

for query in queries: results = library.search_items(query, limit=30) all_results.update(results)

queries = [ "skill transfer", "transfer of learning", "generalization of skills" ]

for query in queries: results = library.search_items(query, limit=30) all_results.update(results)

Rank combined results

对合并结果排序

ranked = orchestrator._rank_items(list(all_results), "skill transfer") print(format_results(ranked[:20]))

undefined

ranked = orchestrator._rank_items(list(all_results), "skill transfer") print(format_results(ranked[:20]))

undefined

Iterative Refinement

迭代优化

python

import sys
sys.path.append('/Users/niyaro/Documents/Code/zotero-code-execution')
import setup_paths
from zotero_lib import ZoteroLibrary, SearchOrchestrator, format_results

library = ZoteroLibrary()
orchestrator = SearchOrchestrator(library)

python

import sys
sys.path.append('/Users/niyaro/Documents/Code/zotero-code-execution')
import setup_paths
from zotero_lib import ZoteroLibrary, SearchOrchestrator, format_results

library = ZoteroLibrary()
orchestrator = SearchOrchestrator(library)

Initial search

初始搜索

initial = library.search_items("memory", limit=50)

Analyze tags

分析标签

tag_freq = {} for item in initial: for tag in item.tags: tag_freq[tag] = tag_freq.get(tag, 0) + 1

Find most common tag

找到最常见的标签

if tag_freq: most_common_tag = max(tag_freq, key=tag_freq.get)

# Refine search
refined = orchestrator.filter_by_criteria(
    initial,
    required_tags=[most_common_tag]
)

print(f"Papers with most common tag '{most_common_tag}':")
print(format_results(refined))

undefined

if tag_freq: most_common_tag = max(tag_freq, key=tag_freq.get)

# 优化搜索
refined = orchestrator.filter_by_criteria(
    initial,
    required_tags=[most_common_tag]
)

print(f"带有最常见标签 '{most_common_tag}' 的论文：")
print(format_results(refined))

undefined

📚 API Reference

📚 API 参考

SearchOrchestrator

SearchOrchestrator

Main class for automated searching.

用于自动化搜索的主类。

comprehensive_search(query, max_results=20, use_semantic=True, use_keyword=True, use_tags=True, search_limit_per_strategy=50)

comprehensive_search(query, max_results=20, use_semantic=True, use_keyword=True, use_tags=True, search_limit_per_strategy=50)

Performs multi-strategy search with automatic deduplication and ranking.

Parameters:

```
query
```
(str): Search query
```
max_results
```
(int): Maximum results to return (default: 20)
```
use_semantic
```
(bool): Use semantic search (default: True)
```
use_keyword
```
(bool): Use keyword search (default: True)
```
use_tags
```
(bool): Use tag search (default: True)
```
search_limit_per_strategy
```
(int): Items to fetch per strategy (default: 50)

Returns: List of ZoteroItem objects

执行多策略搜索，自动完成去重和排序。

参数：

```
query
```
(字符串)：搜索查询词
```
max_results
```
(整数)：返回的最大结果数（默认值：20）
```
use_semantic
```
(布尔值)：是否使用语义搜索（默认值：True）
```
use_keyword
```
(布尔值)：是否使用关键词搜索（默认值：True）
```
use_tags
```
(布尔值)：是否使用标签搜索（默认值：True）
```
search_limit_per_strategy
```
(整数)：每种策略获取的条目数（默认值：50）

返回值： ZoteroItem 对象列表

filter_by_criteria(items, item_types=None, date_range=None, required_tags=None, excluded_tags=None)

filter_by_criteria(items, item_types=None, date_range=None, required_tags=None, excluded_tags=None)

Filter items by various criteria.

Parameters:

```
items
```
(list): Items to filter
```
item_types
```
(list): Allowed item types (e.g., ["journalArticle"])
```
date_range
```
(tuple): (min_year, max_year)
```
required_tags
```
(list): Tags that must be present
```
excluded_tags
```
(list): Tags that must not be present

Returns: Filtered list of ZoteroItem objects

根据多种条件过滤条目。

参数：

```
items
```
(列表)：待过滤的条目
```
item_types
```
(列表)：允许的条目类型（例如：["journalArticle"]）
```
date_range
```
(元组)：(起始年份, 结束年份)
```
required_tags
```
(列表)：必须包含的标签
```
excluded_tags
```
(列表)：必须排除的标签

返回值： 过滤后的ZoteroItem对象列表

ZoteroLibrary

ZoteroLibrary

Low-level interface to Zotero.

与Zotero交互的底层接口。

search_items(query, qmode="titleCreatorYear", item_type="-attachment", limit=100, tag=None)

search_items(query, qmode="titleCreatorYear", item_type="-attachment", limit=100, tag=None)

Basic keyword search.

基础关键词搜索。

semantic_search(query, limit=100, search_type="hybrid")

semantic_search(query, limit=100, search_type="hybrid")

Semantic/vector search.

语义/向量搜索。

search_by_tag(tags, item_type="-attachment", limit=100)

search_by_tag(tags, item_type="-attachment", limit=100)

Search by tags.

基于标签的搜索。

get_recent(limit=50)

get_recent(limit=50)

Get recently added items.

获取最近添加的条目。

get_tags()

get_tags()

Get all tags in library.

获取库中的所有标签。

format_results(items, include_abstracts=True, max_abstract_length=300)

format_results(items, include_abstracts=True, max_abstract_length=300)

Format items as markdown.

将条目格式化为Markdown格式。

⚙️ Configuration

⚙️ 配置

Default Parameters

默认参数

Good defaults for most searches:

python

orchestrator.comprehensive_search(
    query,
    max_results=20,              # Top 20 results
    search_limit_per_strategy=50 # Fetch 50 per strategy
)

适用于大多数搜索场景的默认配置：

python

orchestrator.comprehensive_search(
    query,
    max_results=20,              # 前20条结果
    search_limit_per_strategy=50 # 每种策略获取50条
)

Adjusting Search Depth

调整搜索深度

For quick searches (fewer results, faster):

python

results = orchestrator.comprehensive_search(
    query,
    max_results=10,
    search_limit_per_strategy=20
)

For thorough searches (more comprehensive):

python

results = orchestrator.comprehensive_search(
    query,
    max_results=30,
    search_limit_per_strategy=100
)

快速搜索（结果更少，速度更快）：

python

results = orchestrator.comprehensive_search(
    query,
    max_results=10,
    search_limit_per_strategy=20
)

深度搜索（结果更全面）：

python

results = orchestrator.comprehensive_search(
    query,
    max_results=30,
    search_limit_per_strategy=100
)

🔍 How It Works

🔍 工作原理

Behind the Scenes

幕后流程

When you call

comprehensive_search("embodied cognition", max_results=20)

Semantic Search (if enabled):
- Searches "embodied cognition" (hybrid mode) → 50 items
- Searches "embodied cognition" (vector mode) → 50 items
Keyword Search (if enabled):
- Searches with qmode="everything" → 50 items
- Searches with qmode="titleCreatorYear" → 50 items
Tag Search (if enabled):
- Extracts words from query
- Finds matching tags in library
- Searches by matching tags → 50 items
Processing:
- Combines all results (~250 items)
- Deduplicates using item keys (~120 unique)
- Ranks by relevance score
- Returns top 20
Context:
- Only the final 20 items go to LLM context
- All processing happens in code execution environment

当你调用

comprehensive_search("embodied cognition", max_results=20)

时：

语义搜索（若启用）：
- 搜索“embodied cognition”（混合模式）→ 50条条目
- 搜索“embodied cognition”（向量模式）→ 50条条目
关键词搜索（若启用）：
- 使用qmode="everything"搜索 → 50条条目
- 使用qmode="titleCreatorYear"搜索 → 50条条目
标签搜索（若启用）：
- 从查询词中提取关键词
- 在库中查找匹配的标签
- 按匹配标签搜索 → 50条条目
处理过程：
- 合并所有结果（约250条）
- 通过条目键去重（约120条唯一结果）
- 按相关性得分排序
- 返回前20条结果
上下文处理：
- 仅最终的20条结果进入LLM上下文
- 所有处理操作在代码执行环境中完成

Why This Is Better

优势对比

Old Approach (Direct MCP):

python

undefined

旧方法（直接调用MCP）：

python

undefined

5+ function calls, all results to context

5次以上函数调用，所有结果进入上下文

results1 = zotero_semantic_search("query", limit=10) # Crash risk if > 15 results2 = zotero_search_items("query", limit=10)

results1 = zotero_semantic_search("query", limit=10) # 结果超过15条时有崩溃风险 results2 = zotero_search_items("query", limit=10)

... manual deduplication, no ranking

...手动去重，无排序

All items (50+) load into context

所有条目（50+条）加载到上下文


**New Approach (Code Execution):**
```python


**新方法（代码执行）：**
```python

1 function call, only top results to context

一次函数调用，仅前N条结果进入上下文

results = orchestrator.comprehensive_search("query", max_results=20)

Fetches 250+ items, processes in code, returns top 20

获取250+条条目，在代码中处理，返回前20条

undefined

undefined

🛠️ Error Handling

🛠️ 错误处理

Always handle potential errors:

python

import sys
sys.path.append('/Users/niyaro/Documents/Code/zotero-code-execution')
import setup_paths
from zotero_lib import SearchOrchestrator, format_results

orchestrator = SearchOrchestrator()

try:
    results = orchestrator.comprehensive_search("query", max_results=20)

    if results:
        print(format_results(results))
    else:
        print("No results found. Try a broader search term.")

except Exception as e:
    print(f"Search failed: {e}")
    print("Please check your Zotero MCP configuration.")

始终处理潜在错误：

python

import sys
sys.path.append('/Users/niyaro/Documents/Code/zotero-code-execution')
import setup_paths
from zotero_lib import SearchOrchestrator, format_results

orchestrator = SearchOrchestrator()

try:
    results = orchestrator.comprehensive_search("query", max_results=20)

    if results:
        print(format_results(results))
    else:
        print("未找到结果。请尝试更宽泛的搜索词。")

except Exception as e:
    print(f"搜索失败：{e}")
    print("请检查你的Zotero MCP配置。")

📖 Examples

📖 示例

See

/Users/niyaro/Documents/Code/zotero-code-execution/examples.py

for 8 complete working examples.

查看

/Users/niyaro/Documents/Code/zotero-code-execution/examples.py

获取8个完整的可用示例。

🎓 Quick Reference

🎓 速查参考

Task	Code
Basic search	`orchestrator.comprehensive_search(query, max_results=20)`
Filter by type	`orchestrator.filter_by_criteria(items, item_types=["journalArticle"])`
Filter by date	`orchestrator.filter_by_criteria(items, date_range=(2020, 2025))`
Search author	`library.search_items(author, qmode="titleCreatorYear", limit=50)`
Search by tag	`library.search_by_tag([tags], limit=50)`
Recent items	`library.get_recent(limit=20)`
Format output	`format_results(items, include_abstracts=True)`

任务	代码
基础搜索	`orchestrator.comprehensive_search(query, max_results=20)`
按类型过滤	`orchestrator.filter_by_criteria(items, item_types=["journalArticle"])`
按日期过滤	`orchestrator.filter_by_criteria(items, date_range=(2020, 2025))`
作者搜索	`library.search_items(author, qmode="titleCreatorYear", limit=50)`
标签搜索	`library.search_by_tag([tags], limit=50)`
最近条目	`library.get_recent(limit=20)`
格式化输出	`format_results(items, include_abstracts=True)`

💡 Tips

💡 提示

Start simple: Use
```
comprehensive_search()
```
for most queries
Adjust depth: Use
```
search_limit_per_strategy
```
to control thoroughness
Filter after: Fetch broadly, filter in code
Custom logic: Use Python for complex filtering
Check errors: Always wrap in try/except

从简单开始：大多数查询使用
```
comprehensive_search()
```
即可
调整深度：使用
```
search_limit_per_strategy
```
控制搜索全面性
先获取后过滤：先广泛获取条目，再在代码中过滤
自定义逻辑：使用Python实现复杂过滤
错误检查：始终将代码包裹在try/except块中

📁 Documentation

📁 文档

Quick Start:

/Users/niyaro/Documents/Code/zotero-code-execution/QUICK_START.md

Full Docs:

/Users/niyaro/Documents/Code/zotero-code-execution/README.md

Examples:

/Users/niyaro/Documents/Code/zotero-code-execution/examples.py

Status:

/Users/niyaro/Documents/Code/zotero-code-execution/HONEST_STATUS.md

快速入门：

/Users/niyaro/Documents/Code/zotero-code-execution/QUICK_START.md

完整文档：

/Users/niyaro/Documents/Code/zotero-code-execution/README.md

示例：

/Users/niyaro/Documents/Code/zotero-code-execution/examples.py

状态说明：

/Users/niyaro/Documents/Code/zotero-code-execution/HONEST_STATUS.md

⚠️ Important Notes

⚠️ 重要说明

This uses code execution, not direct MCP calls
Large datasets are processed in code, keeping context small
Semantic search may not be available (falls back to keyword)
Results are automatically deduplicated and ranked
Safe to use large limits (100+) because filtering happens in code

本技能使用代码执行来安全处理大型搜索
大型数据集在代码中处理，保持上下文数据量较小
语义搜索可能不可用（会自动回退到关键词搜索）
结果会自动去重和排序
可安全使用较大的获取上限（100+），因为过滤在代码中完成

🔄 Migration from zotero-mcp

🔄 从zotero-mcp迁移

Old pattern:

python

undefined

旧模式：

python

undefined

Multiple manual MCP calls

多次手动调用MCP

results1 = zotero_semantic_search("query", limit=10) results2 = zotero_search_items("query", limit=10)

Manual deduplication...

手动去重...


**New pattern:**
```python


**新模式：**
```python

One function call with code execution

一次代码执行函数调用

import sys sys.path.append('/Users/niyaro/Documents/Code/zotero-code-execution') import setup_paths from zotero_lib import SearchOrchestrator, format_results

orchestrator = SearchOrchestrator() results = orchestrator.comprehensive_search("query", max_results=20) print(format_results(results))


---

**Remember:** This skill uses code execution to safely handle large searches. The implementation is in `/Users/niyaro/Documents/Code/zotero-code-execution/`.

import sys sys.path.append('/Users/niyaro/Documents/Code/zotero-code-execution') import setup_paths from zotero_lib import SearchOrchestrator, format_results

orchestrator = SearchOrchestrator() results = orchestrator.comprehensive_search("query", max_results=20) print(format_results(results))


---

**注意：** 本技能通过代码执行来安全处理大型搜索，实现代码位于 `/Users/niyaro/Documents/Code/zotero-code-execution/`。