---
canonical_url: "https://docs.composio.dev/toolkits/weaviate_mcp"
markdown_url: "https://docs.composio.dev/toolkits/weaviate_mcp.md"
product: "platform"
version: null
last_updated: null
---

# Weaviate MCP

Ask natural-language questions and run AI-powered searches over your Weaviate Cloud collections through the provider-hosted Query Agent MCP server.

- **Category:** model context protocol
- **Auth:** API_KEY
- **Composio-managed OAuth available?** N/A
- **Tools:** 2
- **Triggers:** 0
- **Slug:** `WEAVIATE_MCP`
- **Version:** 00000000_00

## Tools

### Query agent ask

**Slug:** `WEAVIATE_MCP_QUERY_AGENT_ASK`

Ask Mode transforms your query into actionable searches or
aggregations, and then provides a final answer to the question.

For example, you could ask:

"How many orders related to books were placed last week?"

And the agent will filter for orders, perform semantic search for books
and sort or filter for timestamps from the last week. Then, the agent
will provide a response, answering this question exactly based on the
data retrieved.

The result always carries final_answer. When result_evaluation is
'llm', it additionally carries is_partial_answer (whether information
the query asked for is missing from the answer), missing_information
(what exactly could not be found in the data), and sources (the
retrieved objects the answer actually drew on).

Queries the collections named in the request within your connected Weaviate Cloud cluster. Use collection names from your own cluster; access follows the connected API key permissions.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `query` | string | Yes | Natural-language question to answer from the data. |
| `collections` | array | Yes | Names of the collections to answer from. The collections available in the caller's cluster are listed at the end of this tool's description. |
| `result_evaluation` | string ("llm" | "none") | No | Set to 'llm' to have the agent evaluate its own answer, adding is_partial_answer, missing_information, and sources to the result. With 'none' (the default) the result carries only final_answer. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `sources` | array | No |  |
| `final_answer` | string | Yes |  |
| `is_partial_answer` | boolean | No |  |
| `missing_information` | array | No |  |

### Query agent search

**Slug:** `WEAVIATE_MCP_QUERY_AGENT_SEARCH`

Search Mode combines AI-powered semantic search with structured
filtering and returns the matching Weaviate objects directly.

For example, you could ask:

"Find me some vintage shoes under $70"

And the agent will perform semantic search for vintage shoes, apply a
filter for price < 70, and return the matching objects from your
collections, ready for you to render or post-process.

You could also ask:

"Something comfortable to wear on a long flight"

And the agent will use AI-powered search to find relevant objects, even
when terms like comfortable or long flight never appear in your data.

Under the hood, Search Mode does more than embed your query as-is. The
agent writes one or more optimized semantic and structured queries,
executes them against your collections, and reranks the retrieved
objects by how well each one matches your original request.

The result's `results` field lists the matching objects in rank order
(best match first), each as a dict of property names to values.

Queries the collections named in the request within your connected Weaviate Cloud cluster. Use collection names from your own cluster; access follows the connected API key permissions.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `limit` | integer | No | Maximum results to return. |
| `query` | string | Yes | Natural-language search query. |
| `effort` | string ("medium" | "high" | "ultrahigh") | No | Search effort: 'medium', 'high', or 'ultrahigh'. Higher effort reranks more candidates. |
| `offset` | integer | No | Results to skip, for pagination. |
| `filtering` | string ("precision" | "recall") | No | 'precision' applies the filters the agent writes strictly; 'recall' (the default) hedges uncertain filters by also searching without them, so plausible matches are not filtered out. |
| `collections` | array | Yes | Names of the collections to search. The collections available in the caller's cluster are listed at the end of this tool's description. |
| `diversity_weight` | string | No | Trade relevance against result diversity: 0 is pure relevance, 1 is maximum diversity. Requires all queried collections to use a supported vectorizer; leave unset to rank purely by relevance. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `results` | array | Yes |  |