Weaviate Launches Agent Skills to Empower AI Coding Agents
Amsterdam, Netherlands, Feb. 21, 2026 (GLOBE NEWSWIRE) — February 20, 2026 – Weaviate, the main open-source AI database, right now introduced the launch of Weaviate Agent Skills, an revolutionary open-source repository that equips common coding brokers like Claude Code, Cursor, GitHub Copilot, VS Code, and Gemini CLI with exact instruments for producing production-ready code focusing on Weaviate workflows.
This launch builds instantly on Weaviate’s Question Agent, first previewed in March 2025 and reaching normal availability in September 2025. The Question Agent helps pure language queries throughout a number of collections, that includes multi-collection routing, clever question growth, decomposition for complicated questions, user-defined filters, and reranking for optimum outcomes. Builders can take a look at Agent Abilities instantly utilizing Weaviate Cloud’s free Sandbox clusters—small cases designed for experimentation that final 14 days and could be prolonged or upgraded to manufacturing Shared Cloud setups.
Complete Repository Instruments
The repository at github.com/weaviate/agent-skills is structured into two core sections, offering full lifecycle help from fundamental operations to finish functions.
Weaviate Abilities within the /expertise/weaviate listing provide granular scripts for key duties. These cowl cluster administration equivalent to schema inspection, assortment creation, and metadata retrieval; knowledge lifecycle operations together with imports from CSV, JSON, or JSONL recordsdata plus instance knowledge era; agentic search powered by Question Agent; and superior retrieval choices like hybrid search (mixing semantic and key phrase with alpha parameters), pure semantic, or key phrase modes.
Cookbooks within the /expertise/weaviate-cookbooks folder present end-to-end blueprints for manufacturing apps. Highlights embrace Question Agent chatbots constructed with FastAPI backends and Subsequent.js frontends; multimodal PDF RAG pipelines utilizing ModernVBERT for multivector embeddings alongside Ollama or Qwen3-VL for era; fundamental, superior, and agentic RAG implementations with decomposition and reranking; and DSPy-optimized brokers incorporating customized instruments and chronic reminiscence.
Six Streamlined Slash Instructions
Agent Abilities introduces six intuitive instructions that AI coding brokers can auto-discover and execute, streamlining Weaviate interactions:
- /weaviate:ask: Delivers AI-generated solutions with citations through Question Agent.
- /weaviate:collections: Lists all schemas or inspects particular collections.
- /weaviate:discover: Reveals knowledge metrics, counts, and pattern objects.
- /weaviate:fetch: Retrieves objects by ID or filters by properties.
- /weaviate:question: Performs pure language searches throughout collections.
- /weaviate:search: Executes hybrid, semantic, or key phrase searches with parameters like alpha mixing.
As an illustration, builders can run “/weaviate:search question ‘greatest laptops’ assortment ‘Merchandise’ sort ‘hybrid’ alpha ‘0.7’” for balanced retrieval or “/weaviate:ask What are vector database advantages?” in opposition to a Documentation assortment.
CEO Bob van Luijt’s Imaginative and prescient
Bob van Luijt, Co-Founder and CEO of Weaviate—which he launched as an open-source vector search engine in March 2019—shared launch insights. “Weaviate Agent Abilities bridges the hole between high-velocity AI coding and dependable infrastructure, letting builders construct refined AI techniques with out debugging agent hallucinations,” van Luijt said.
As a outstanding Netherlands-based know-how entrepreneur, Van Luijt champions open-source AI instruments. He positions Weaviate as a “batteries-included” stack that mixes vector search, structured filtering, and agentic capabilities for contemporary AI functions.
On the spot Setup for Builders
Integration is designed for velocity. Set up with a single command like npx expertise add weaviate/agent-skills or through plugin managers in instruments like Claude Code. Configure atmosphere variables utilizing your Weaviate Cloud endpoint and API key from a free Sandbox cluster.
Run /weaviate:quickstart for guided setup. This launch aligns with Weaviate’s 2025 momentum, together with Question Agent GA, enhanced TypeScript/Python SDKs, multi-turn conversations, streaming responses, and new C#/Java shoppers for broader ecosystem help.
Weaviate invitations the group to star the repo, submit pull requests for brand spanking new cookbooks, and take part in discussions on GitHub, the Weaviate Discussion board, Slack workspace, and X.
Strategic Influence on AI Improvement
Agent Abilities addresses a crucial ache level: AI brokers typically generate inaccurate or incomplete code for vector databases on account of hallucinations or outdated data. By offering verified, modular instruments, Weaviate permits quicker iteration from prototype to manufacturing.
Early adopters report 3x reductions in debugging time for RAG pipelines and agentic apps. The repository’s modular design additionally facilitates contributions, with plans for expanded expertise overlaying generative modules, tenancy isolation, and hybrid cloud deployments.
About Weaviate
Weaviate is an open-source, AI database that handles storage, retrieval, and orchestration for generative AI at scale. Backed by enterprise-grade Weaviate Cloud companies, it powers agentic workflows—from easy semantic search to complicated multi-agent techniques—delivering sub-second latency on billions of objects.
Media Contact:
Philip Vollet
PR@weaviate.io
+49-160-96488554
