Intent shortcut: nearText search with limit 3 for tight RAG snippets. Pass className + concepts JSON array. Use WV_NEAR_TEXT_SEARCH when you need custom limits.
Generate MCP URLOverview: Intent shortcut: nearText search with limit 3 for tight RAG snippets. Pass className + concepts JSON array. Use WV_NEAR_TEXT_SEARCH when you need custom limits through Weaviate.
Benefits:
Common Use Cases:
1. Semantic Knowledge Base Search
AI agents query Weaviate to power natural language search across company knowledge bases, returning contextually relevant results with generative summaries.
2. RAG Pipeline with Generative Modules
AI agents leverage Weaviate's generative search modules in retrieval-augmented generation workflows, combining vector retrieval with LLM-powered answer synthesis.
3. Multi-Modal Data Retrieval
AI agents store and query text, image, and audio embeddings in Weaviate, enabling cross-modal search and discovery across diverse content types.
4. Automated Data Classification
AI agents use Weaviate's classification modules to automatically categorize incoming data � tickets, documents, leads � based on semantic similarity to labeled examples.
5. Real-Time Recommendation Engine
AI agents query Weaviate embeddings to deliver personalized content, product, or resource recommendations to users based on behavioral and contextual signals.
Hybrid search preset with alpha=0.75 favoring vectors. Same as WV_HYBRID_SEARCH but encodes a common RAG default. Override with WV_HYBRID_SEARCH for full control.
Readiness GET /.well-known/ready on root. Returns when API can serve. Pair with WV_HEALTH_LIVE.
Builds a nearText Get query for class + concepts + limit + optional where filter. Returns GraphQL data. Use when using text2vec modules; for raw vectors use WV_NEAR_VECTOR_SEARCH.
Checks if a class is defined by calling WV_GET_CLASS and mapping to {exists:boolean}. Use in deployment scripts; for objects use WV_EXISTS_OBJECT.
Updates class metadata PUT /v1/schema/{className}. Returns updated class. Use for inverted index toggles; destructive removal uses WV_DELETE_CLASS.
Gets object by id GET /v1/objects/{id}?include=vector. Returns properties and optional vector. Use WV_LIST_OBJECTS to browse without id.
Starts backup POST /v1/backups/{backend} with id and include classes list. Returns async backup id. Use filesystem or s3 backend string; poll WV_GET_BACKUP_STATUS.
Adds a cross-reference POST /v1/objects/{id}/references/{prop}. Returns beacon info. Use for graph links; delete with WV_DELETE_REFERENCE.
Shortcut Aggregate total count for a class: builds `{Aggregate{ClassName(meta{count})}}`. Returns count only. Use WV_AGGREGATE_QUERY for groupBy and metrics.
Derives enabled modules from WV_GET_META.modules field. Returns module map. Use to see text2vec-openai/cohere availability.
Runs arbitrary GraphQL POST /v1/graphql. Supports Get with nearText, nearVector, hybrid, bm25, Aggregate. Example: {Get{Article(nearText:{concepts:} limit:5){title}}}. Most powerful search entry point.
Creates tenants POST /v1/schema/{className}/tenants with body array. Returns status. Use for SaaS isolation; delete via WV_DELETE_TENANTS.
Returns the full Weaviate class schema via GET /v1/schema. Lists classes, properties, and vectorizer config. Use before WV_CREATE_CLASS; for one class use WV_GET_CLASS.
Cluster node status GET /v1/nodes. Returns shards and health per node. Use on multi-node installs; single docker may return one node.
Replaces object PUT /v1/objects/{id}. Returns updated object. Use for full document replace; partial merges use WV_PATCH_OBJECT.
GraphQL hybrid nearText + BM25 with alpha (0=keyword,1=semantic). Returns ranked objects. Use for best-effort RAG quality; pure keyword use WV_BM25_SEARCH.
Patches properties PATCH /v1/objects/{id}. Returns updated fields. Use for partial updates; full replace uses WV_UPDATE_OBJECT.
Deletes a reference DELETE /v1/objects/{id}/references/{prop}. Returns status. Use to unlink without deleting objects.
Lists objects GET /v1/objects with class, limit, after cursor pagination. Returns ids and summaries. Paginate with after from last result.
Deletes tenants DELETE /v1/schema/{className}/tenants with body listing tenant names. Returns status. Irreversible for tenant data.
Liveness GET /.well-known/live on Weaviate root (non-/v1). Returns plain ok when process alive. Use in k8s probes.
Deletes a class and its objects DELETE /v1/schema/{className}. Irreversible. Export data first; use WV_UPDATE_CLASS for non-destructive changes.
Shortcut list with limit default 25 and class required for common browsing. Returns first page only; follow WV_LIST_OBJECTS with after cursor.
Lists multi-tenancy tenants GET /v1/schema/{className}/tenants. Returns tenant ids and activity. Use before WV_CREATE_TENANTS.
GraphQL Explore across classes with nearVector (optional class filter in query string you provide). Pass full Explore {...} in query param for flexibility.
Fetches a single class definition GET /v1/schema/{className}. Returns properties and moduleConfig. Use when tuning vectorizers; list all with WV_GET_SCHEMA.
Batch add references POST /v1/batch/references. Returns per-item status. Use after objects exist; mirrors WV_ADD_REFERENCE at scale.
Lists backups GET /v1/backups/{backend}. Returns ids. Paginate if many; use WV_GET_BACKUP_STATUS for detail.
BM25 keyword Get query with query text and optional properties list. Returns keyword-ranked objects. Use for lexical recall; semantic similarity uses WV_NEAR_TEXT_SEARCH.
Creates one object POST /v1/objects with class, properties, optional vector. Returns id and beacon. Use WV_BATCH_CREATE_OBJECTS for throughput.
Batch create POST /v1/batch/objects with objects array and optional consistency_level. Returns results per item. Use for imports; single creates use WV_CREATE_OBJECT.
Creates class then polls WV_GET_META until class appears in schema or timeout (best-effort). Returns final schema fragment. Use in automation; interactive use WV_CREATE_CLASS.
GraphQL Get with where filter JSON and limit (no vector). Returns matching objects. Example where: {path:,operator:Equal,valueText:"active"}. Vector search should use WV_NEAR_TEXT_SEARCH.
Adds a property POST /v1/schema/{className}/properties. Returns updated class. Use when evolving schema; reindex may be required for some module changes.
Deletes object DELETE /v1/objects/{id}. Returns status. Irreversible; versions may apply if enabled.
Deletes objects matching where filter POST /v1/batch/objects/delete. Returns match counts. Destructive bulk op; prefer WV_DELETE_OBJECT for singles.
Restores from backup POST /v1/backups/{backend}/{backupId}/restore. Returns restore job info. Destructive to target classes—plan carefully.
GraphQL Get with nearVector for class + vector array + limit. Returns hits with _additional.distance. Use when you already embedded the query; concepts-based search uses WV_NEAR_TEXT_SEARCH.
Updates tenant activity POST /v1/schema/{className}/tenants (PUT/PATCH per server version—body includes name + activityStatus). Returns updated tenant. Use to freeze tenants.
Reads cluster meta GET /v1/meta (version, modules). Use for compatibility checks; process health uses WV_HEALTH_LIVE.
HEAD /v1/objects/{id} to test existence without a body. Returns success headers only. Use before conditional updates.
Gets backup status GET /v1/backups/{backend}/{backupId}. Returns progress. Use after WV_CREATE_BACKUP; restore with WV_RESTORE_BACKUP.
Creates a class POST /v1/schema with class, vectorizer, moduleConfig, properties. Returns schema object. Use WV_CREATE_CLASS_AND_WAIT if you need to poll readiness in scripts.
GraphQL Aggregate for counts, grouped metrics (sum/mean) on a class. Returns facet statistics. Use for analytics; row-level retrieval uses WV_GRAPHQL_QUERY Get.
Do I need my own developer credentials to use Weaviate MCP with Adopt AI?
No, you can get started immediately using Adopt AI's built-in Weaviate integration. For production use, we recommend configuring your own API credentials for greater control and security.
Can I connect Weaviate with other apps through Adopt AI?
Yes! Adopt AI supports multi-app workflows, so your AI agents can seamlessly move data between Weaviate and LLM platforms, knowledge bases, search tools, and more.
Is Adopt AI secure?
Absolutely. Adopt AI is SOC 2 Type 2 certified and ISO/IEC 27001 compliant, and adheres to EU GDPR, CCPA, and HIPAA standards. All data is encrypted in transit and at rest, ensuring the confidentiality, integrity, and availability of your data. Learn more here.
What happens if the Weaviate API changes?
Adopt AI maintains and updates all integrations automatically, so your agents always work with the latest API versions, no manual maintenance required.
Do I need coding skills to set up the Weaviate integration?
Not at all. Adopt AI's zero-shot API discovery means your agents understand Weaviate's schema on first contact. Setup takes minutes with no code required.
How do I set up custom Weaviate MCP in Adopt AI?
For a step-by-step guide on creating and configuring your own Weaviate API credentials with Adopt AI, see here.
