Connect AI assistants to approved enterprise data through one approved MCP server. Credentials stay in MarcoPolo, data work runs outside the model, and every action is logged.
Connect assistants to enterprise applications and databases through one governed MCP gateway. Assistants see only the connections and actions they’re approved to use, while MarcoPolo handles credentials, execution, results, and audit.
Claude
ChatGPT
CursorCredentials resolved here · never in model context
Salesforcesales_read
Snowflakeanalytics_ro
Postgresapp_read
Jiraissues_readAssistants discover visible, ready connections and the capabilities policy allows. Credentials stay in MarcoPolo. The model sees the tool, not the token.
Secure connections# The assistant calls a tool, not the database.▸ tool marcopolo.run_query→ salesforce_opportunities 18 rows→ jira_issues 64 rows✓ slipped_deals_joined 14 rows▸ tool marcopolo.preview# The model reads 14 rows · not 82.▍
Assistants can run approved data queries and workspace commands, receive execution status, preview bounded results, and continue from result handles.
Governed workspaceEvery assistant call lands in the same trail
Every tool call carries identity, tenant, connection, execution, result, and effect metadata. Assistant activity lands in the same audit trail as the rest of the gateway.
Trust and governanceGive assistants a small set of governed tools instead of a pile of provider-specific integrations. The assistant asks for data work, and MarcoPolo handles access, execution, results, and audit.
People provision credentials, approve grants, and decide which connections and operations are available.
Assistants receive a small set of tools for discovery, execution, result preview, and guided continuation.
When a team needs product-specific orchestration, the same gateway is available through direct SDK calls.
MCP calls create execution and audit records alongside SDK and web app activity.
An analyst asks an assistant to investigate renewal risk. The assistant uses approved CRM and support connections, runs data work in MarcoPolo, and returns a result-backed summary.
A model finds a useful result through MCP, then a custom application uses the SDK to continue the workflow with the same result handle and tenant context.
A provider failure returns a failed execution with structured recovery guidance, rather than leaving the assistant to guess what happened.
No. MCP is for assistants. The SDK is for application developers. Both use the same gateway.
No. MarcoPolo MCP sits above the connection layer, so assistants use one governed tool surface across approved systems.
Credential setup and sharing are better handled through the web app or explicit SDK workflows. MCP should expose the runtime capabilities that policy allows.
Same data gateway, same governance model. What changes is who is driving and where the work is written.
Governed connections, managed execution, and audit inside the agent you ship.
Read the Agent SDK pageAgent ConsoleProvision access, review what agents did, and hold the controls in one place.
Read the Agent Console pageLiveConnect100+ live enterprise systems behind the same six verbs, scoped per user and agent.
Read the LiveConnect pageAdaptive ContextA knowledge map per connection: schema, business rules, and joins that already worked.
Read the Adaptive Context page
Give MCP-compatible clients governed tools for approved connections, managed data execution, and auditable results.
