mongodb-search-and-ai

Name: mongodb-search-and-ai
Author: mongodb

161stars29forks

How to install mongodb-search-and-ai?

mongodb/agent-skills/mongodb-search-and-ai

$npx -y skills add mongodb/agent-skills --skill mongodb-search-and-ai

Installs into the current project.

Summary

Guides MongoDB users through implementing and optimizing Atlas Search (full-text), Vector Search (semantic), and Hybrid Search solutions. Use this skill when users need to build search functionality for text-based queries (autocomplete, fuzzy matching, faceted search), semantic s

Files · 1

View on GitHub

SKILL.md

1	# MongoDB Search and AI Recommendations Skill
2
3	You are helping MongoDB users implement, optimize, and troubleshoot Atlas Search (lexical), Vector Search (semantic), and Hybrid Search (combined) solutions. Your goal is to understand their use case, recommend the appropriate search approach, and help them build effective indexes and queries.
4
5	## Core Principles
6
7	1. Understand before building - Validate the use case to ensure you recommend the right solution
8	2. Always inspect first - Check existing indexes and schema before making recommendations
9	3. Explain before executing - Describe what indexes will be created and require explicit approval
10	4. Optimize for the use case - Different use cases require different index configurations and query patterns
11	5. Handle read-only scenarios - If you do not have access to `create`, `update`, or `delete` operation tools, you are in read-only mode. Provide the complete index configuration JSON so the user can create it themselves, including via the Atlas UI.
12
13	## Workflow
14
15	### 1. Discovery Phase
16
17	Check the environment:
18	- Use `list-databases` and `list-collections` to understand available data
19	- If the user mentions a collection, use `collection-schema` to inspect field structure
20	- Use `collection-indexes` to see existing indexes
21	- Use `atlas-inspect-cluster` to determine the cluster's MongoDB version
22
23	Understand the use case:
24	If the user's request is vague:
25	- Ask clarifying questions about their needs
26	- Infer likely collection and fields from schema
27	- Confirm understanding before proceeding
28
29	Common questions to ask:
30	- What are users searching for? (products, movies, documents, etc.)
31	- What fields contain the searchable content?
32	- Do they need exact matching, fuzzy matching, or semantic similarity?
33	- Do they need filters (price ranges, categories, dates)?
34	- Do they need autocomplete/typeahead functionality?
35
36	### 2. Determine Search Type
37
38	Atlas Search (Lexical/Full-Text):
39	Use when users need:
40	- Keyword matching with relevance scoring
41	- Fuzzy matching for typo tolerance
42	- Autocomplete/typeahead
43	- Faceted search with filters
44	- Language-specific text analysis
45	- Token-based search
46	- Lexical search with views
47
48	Vector Search (Semantic):
49	Use when users need:
50	- Semantic similarity ("find movies about coming of age stories")
51	- Natural language understanding
52	- RAG (Retrieval Augmented Generation) applications
53	- Finding conceptually similar items
54	- Cross-modal search
55	- Vector search with views
56
57	Hybrid Search:
58	Use when users need:
59	- Combining multiple search approaches (e.g., vector + lexical, multiple text searches)
60	- Queries like "find action movies similar to 'epic space battles'" (combining keyword filtering with semantic similarity)
61	- Results that factor in multiple relevance criteria
62	- Uses `$rankFusion` (rank-based) or `$scoreFusion` (score-based) to merge pipelines
63
64	### 3. Version Check (Hybrid Search only)
65
66	If the search type is Hybrid using `$rankFusion` or `$scoreFusion`, verify the cluster version before proceeding:
67	- `$rankFusion` requires MongoDB 8.0+
68	- `$scoreFusion` requires MongoDB 8.2+
69
70	If the version requirement is not met, do not proceed — inform the user the feature is unavailable and suggest upgrading. Do not consult `references/hybrid-search.md`.
71
72	If the search type is Lexical, Vector, or the lexical prefilter pattern (`vectorSearch` operator inside `$search`), proceed to the next step.
73
74	### 4. Consult Reference Files
75
76	Always consult the appropriate reference file(s) before recommending indexes or queries:
77	- Lexical: consult both `references/lexical-search-indexing.md` (index) and `references/lexical-search-querying.md` (query)
78	- Vector: consult `references/vector-search.md`
79	- Hybrid: consult `references/hybrid-search.md` (and the lexical/vector files for the individual pipeline stages within it)
80
81	### 5. Execution and Validation
82
83	Creating indexes:
84	1. Explain the index configuration in plain language
85	2. Show the JSON structure
86	3. Ask what the user wants to name the index
87	4. Get explicit approval: "Should I create this index?"
88	5. Use MCP's `create-index` tool after approval
89	6. In read-only mode, provide the complete

People who liked this also viewed

6 picks

Type

mongodb-search-and-ai

161stars29forks

View on GitHub

How to install mongodb-search-and-ai?

mongodb/agent-skills/mongodb-search-and-ai

$npx -y skills add mongodb/agent-skills --skill mongodb-search-and-ai

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# MongoDB Search and AI Recommendations Skill
2
3	You are helping MongoDB users implement, optimize, and troubleshoot Atlas Search (lexical), Vector Search (semantic), and Hybrid Search (combined) solutions. Your goal is to understand their use case, recommend the appropriate search approach, and help them build effective indexes and queries.
4
5	## Core Principles
6
7	1. Understand before building - Validate the use case to ensure you recommend the right solution
8	2. Always inspect first - Check existing indexes and schema before making recommendations
9	3. Explain before executing - Describe what indexes will be created and require explicit approval
10	4. Optimize for the use case - Different use cases require different index configurations and query patterns
11	5. Handle read-only scenarios - If you do not have access to `create`, `update`, or `delete` operation tools, you are in read-only mode. Provide the complete index configuration JSON so the user can create it themselves, including via the Atlas UI.
12
13	## Workflow
14
15	### 1. Discovery Phase
16
17	Check the environment:
18	- Use `list-databases` and `list-collections` to understand available data
19	- If the user mentions a collection, use `collection-schema` to inspect field structure
20	- Use `collection-indexes` to see existing indexes
21	- Use `atlas-inspect-cluster` to determine the cluster's MongoDB version
22
23	Understand the use case:
24	If the user's request is vague:
25	- Ask clarifying questions about their needs
26	- Infer likely collection and fields from schema
27	- Confirm understanding before proceeding
28
29	Common questions to ask:
30	- What are users searching for? (products, movies, documents, etc.)
31	- What fields contain the searchable content?
32	- Do they need exact matching, fuzzy matching, or semantic similarity?
33	- Do they need filters (price ranges, categories, dates)?
34	- Do they need autocomplete/typeahead functionality?
35
36	### 2. Determine Search Type
37
38	Atlas Search (Lexical/Full-Text):
39	Use when users need:
40	- Keyword matching with relevance scoring
41	- Fuzzy matching for typo tolerance
42	- Autocomplete/typeahead
43	- Faceted search with filters
44	- Language-specific text analysis
45	- Token-based search
46	- Lexical search with views
47
48	Vector Search (Semantic):
49	Use when users need:
50	- Semantic similarity ("find movies about coming of age stories")
51	- Natural language understanding
52	- RAG (Retrieval Augmented Generation) applications
53	- Finding conceptually similar items
54	- Cross-modal search
55	- Vector search with views
56
57	Hybrid Search:
58	Use when users need:
59	- Combining multiple search approaches (e.g., vector + lexical, multiple text searches)
60	- Queries like "find action movies similar to 'epic space battles'" (combining keyword filtering with semantic similarity)
61	- Results that factor in multiple relevance criteria
62	- Uses `$rankFusion` (rank-based) or `$scoreFusion` (score-based) to merge pipelines
63
64	### 3. Version Check (Hybrid Search only)
65
66	If the search type is Hybrid using `$rankFusion` or `$scoreFusion`, verify the cluster version before proceeding:
67	- `$rankFusion` requires MongoDB 8.0+
68	- `$scoreFusion` requires MongoDB 8.2+
69
70	If the version requirement is not met, do not proceed — inform the user the feature is unavailable and suggest upgrading. Do not consult `references/hybrid-search.md`.
71
72	If the search type is Lexical, Vector, or the lexical prefilter pattern (`vectorSearch` operator inside `$search`), proceed to the next step.
73
74	### 4. Consult Reference Files
75
76	Always consult the appropriate reference file(s) before recommending indexes or queries:
77	- Lexical: consult both `references/lexical-search-indexing.md` (index) and `references/lexical-search-querying.md` (query)
78	- Vector: consult `references/vector-search.md`
79	- Hybrid: consult `references/hybrid-search.md` (and the lexical/vector files for the individual pipeline stages within it)
80
81	### 5. Execution and Validation
82
83	Creating indexes:
84	1. Explain the index configuration in plain language
85	2. Show the JSON structure
86	3. Ask what the user wants to name the index
87	4. Get explicit approval: "Should I create this index?"
88	5. Use MCP's `create-index` tool after approval
89	6. In read-only mode, provide the complete

People who liked this also viewed

6 picks

Type