OpenAI 
Anthropic 
Azure OpenAI 
Ollama 
IBM Db2 
PostgreSQL 
Oracle Choose the model estate
OpenAI, OpenAI-compatible endpoints, Anthropic, Azure OpenAI, or local Ollama. Provider credentials stay encrypted in your Core Hub.
Gluesync AI Hub · Available in 2.3
The agent platform inside your data control plane.
Build and run Spark agents with your models, Core Hub tools, and existing permissions—grounded in an Enterprise brain that every AI tool your company uses can ask, from chat, SQL, APIs, automations, and the clients your team already uses.
New in AI Hub · Enterprise brain
Answers without waiting for the next BI view.
Executives can finally get a tool that answers questions without waiting months for a new BI view or another spreadsheet estate. The Enterprise brain begins where Gluesync is already strong, integrating databases, and extends without limits through the skills and company tools you plug in while in platform.
See it as the foundation for all of your AI tools. Grok bots, Meta Muse, Claude, ChatGPT, Hermes Agent, OpenClaw: they can all talk to the Gluesync Enterprise Brain and get immediate answers based on company shared knowledge. You own it, you control it, it is not SaaS. It is yours, under your own rules.
- Schema graphEvery connected database, one graph of tables, columns, and keys
- Shared memoryRoutes and procedures remembered for the next person, never rows or secrets
- SkillsAgents publish what they learn as immutable, versioned skills
- Company toolsGitLab, Notion, and any approved HTTPS MCP server, refused everywhere else
- Any AI clientClaude, ChatGPT, Grok, Meta Muse, Hermes Agent, OpenClaw, Cursor
- YoursA dedicated brain database on your infrastructure, fed by your own models
From model access to production work
More than a chat surface.
AI Hub is the shared model, agent, execution, and governance layer behind every Gluesync AI experience.
Publish what the agent is
Freeze instructions, skill versions, routing policy, limits, and output contracts into an immutable Spark agent version.
Run work durably
Create asynchronous runs with idempotency, cancellation, correlation, redacted outcomes, and ordered SSE event history.
Bound access and spend
Use scoped platform keys, agent and model allow-lists, rate limits, expiry, budgets, usage, and safe audit records.
AI Studio Query Studio Query Forge SQL Chronos Native API OpenAI SDKs MCP clients
One platform · Multiple entry points
Meet each team where its work already happens.
01
One brain for every AI tool
Enterprise brain
Turn the databases you already integrate into one shared schema graph and memory that Spark agents, Claude, ChatGPT, Grok, Meta Muse, and any MCP client can ask. Self-hosted, shared by default, private on request.
Explore → 02Build and operate agents
AI Studio
Chat with Spark agents, publish immutable versions and skills, run durable jobs, route models, govern spend, and expose the platform through scoped APIs.
Explore → 03Ask in SQL, review the SQL
SQL AI
Query Studio AI drafts and completes statements from the schema it already knows. The gluesync.ai virtual table lets any JDBC client, BI tool, or batch job ask a question and read the answer, a proposed SQL draft, and a conversation ID back as one row.
Explore → 04Agents on schedules, events, and signals
AI Workflows
Fire published Spark agent versions from Chronos schedules, Core Hub platform events, webhooks, chained pipeline steps, your applications, or SQL. Durable, idempotent, budgeted runs with payload allow-lists, loop guards, and approvals on every write.
Explore → 05What teams do with it
AI use cases
Ask your data, triage sync health, explain alerts, discover PII, stand up pipelines from a conversation, validate migrations, embed agents in applications, and answer executive questions from the AI client the company already uses.
Explore → 06One tool catalog
Core Hub MCP
Give in-product Spark agents and external MCP clients access to 125 Core Hub tools, including Validator and the Enterprise brain, under the caller’s existing token and permissions.
Explore → 07Reach the record
AI on legacy databases
Bring IBM i, Db2, Oracle, and other established systems into governed AI workflows through the Gluesync agents already beside them.
Explore → 08Meet Spark
Spark
Spark is Gluesync’s AI agent layer. It grounds every turn in the Enterprise brain, then uses Core Hub MCP tools, MOLO17’s online MCP tools for documentation and knowledge, and Gluesync Connect capabilities.
Explore →Gluesync AI Hub
Your model can change. Your control plane stays in charge.
- Your infrastructure
- Your models
- Your permissions
- Read-only defaults
- Usage & audit
AI Hub is available in Gluesync 2.3. Start with AI Studio for in-product Spark agents, or connect the broader Gluesync platform that supplies their operational and data context.