Definition

What is Wire?

Last updated

A Context as a Service platform that lets you create composable context containers accessible to AI agents through MCP.

You create containers, upload files or let agents write entries directly, and Wire processes everything into AI-optimized context with built-in MCP tools. Teams use Wire to share context across all their AI tools without duplication or configuration overhead.

  • Organizes context into containers: isolated, permissioned environments that agents access through MCP or a REST API.
  • Every container exposes five standard tools: wire_explore, wire_navigate, wire_search, wire_write, and wire_delete.
  • Processes uploaded files and agent-written entries into structured context with relationships and embeddings.
  • Containers are private by default and shareable across a team with role-based access.
  • Context stays portable: the same container serves every MCP-capable tool, so what one agent learns is available to the next.
Related Context as a Service Context Container MCP (Model Context Protocol)

How Wire works

You create a container, then add context to it in two ways: upload files, or let agents write entries directly with wire_write. Wire processes whatever comes in, extracting structure, discovering relationships between entries, and generating embeddings, so the content becomes queryable context rather than a pile of documents.

Agents access the container through its own MCP server or the REST API. The five standard tools cover the whole lifecycle: wire_explore and wire_navigate for orientation and browsing, wire_search for retrieval, wire_write for adding entries, and wire_delete for removing them. An agent connecting for the first time can find its way around a container it has never seen.

Why Wire matters

Most teams now use several AI tools, and each one starts from zero. Context explained to one assistant is invisible to the next, which is the AI silo problem: the more tools you use, the more copies of your context you maintain, and the staler each copy gets.

Wire externalizes context instead. Containers hold it once, agents query it at runtime, and updates are immediately visible to every connected tool. Context becomes something you own and carry between tools rather than something each tool accumulates and locks in, the practical version of moving your AI context between tools.

Common misconceptions about Wire

  • “Wire is file storage with an API.” Files are one input, not the product. Uploads are processed into structured entries with relationships and embeddings, and agents query that structure rather than re-reading documents.
  • “Wire is an agent memory plugin.” Persistent memory is one use case, and wire-memory is built on containers. But containers hold any context a team wants agents to share: documentation, research, project state, customer data.
  • “Connecting another agent means another integration.” A container’s MCP server serves every MCP-capable client the same way. Adding a new tool to your stack means pointing it at the same URL, not building a new sync.

Wire and MCP

MCP standardizes how agents call tools; Wire supplies the context behind those tools. Every container is an MCP server with the five standard tools, so the protocol handles the connection and Wire handles ingestion, structure, retrieval, and permissions. For agents, a Wire container is simply an MCP server that happens to know your context.

FAQ

Frequently asked questions

Common questions about Wire.

Is Wire a vector database?
No. A vector database is one piece of retrieval infrastructure you run and wire up yourself. Wire is the context layer above that: it ingests files and entries, structures them, handles retrieval, and exposes the result to agents as ready-made tools with access control built in.
How do agents connect to Wire?
Each container runs its own MCP server, so any MCP-capable agent connects with a URL and gets the container's tools immediately. There is also a REST API for applications that want direct programmatic access.
Do I need an account to try Wire?
No. Agents can connect to an ephemeral container without a signup and start reading and writing context right away. Ephemeral containers can be claimed into a permanent account later, keeping everything already stored in them.
Is my data private?
Containers are private by default. Sharing is explicit and scoped: organization members get role-based access, and connected agents only see the containers they are granted. Deleting a container deletes its data, including derived artifacts like embeddings and search indexes.
What is the difference between Wire and a RAG framework?
A RAG framework is code you assemble and operate inside one application. Wire is a managed context layer that any number of agents and tools share: storage, processing, retrieval, and permissions are already built, and agents consume them through standard tools instead of custom pipeline code.

All terms

View full glossary
Agent Drift AI Agent Agent Reliability AI Hallucination AI Second Brain Chunking Context as a Service Context Compression Context Bloat Context Container Context Engineering Context Offloading Context Drift Context Portability Context Pruning Context Poisoning Context Rot Context Window Epistemic Provenance Fine-Tuning Knowledge Graph MCP Resources MCP Server MCP (Model Context Protocol) Multi-Agent System Prompt Caching Prompt Engineering Semantic Search RAG (Retrieval-Augmented Generation) Structured Context Tool Poisoning Wire

Put context into practice

Create your first context container and connect it to your AI tools in minutes.

Create Your First Container