Connic
Back to BlogChangelog

What We Shipped in November 2025

MCP connector exposing agents as tools, Postgres LISTEN/NOTIFY, S3 file uploads, SQS message queues, connector logs, and unified connector UI.

December 2, 20255 min read

November was connector month with four new integrations covering databases, cloud storage, message queues, and the Model Context Protocol.

MCP Connector

Expose your Connic agents as MCP tools that any MCP client can call. Each linked agent becomes a tool with its name and description available via the standard MCP protocol.

The connector supports two modes:

  • Sync mode: MCP client waits for the agent to complete (up to 5 minutes)
  • Inbound mode: Fire-and-forget for longer-running tasks

Connect Claude Desktop, Cursor, or any MCP-compatible client to your Connic agents. Authentication is handled via a generated secret token.

Postgres Connector

Trigger agents from database events using PostgreSQL's LISTEN/NOTIFY mechanism. Configure a channel name, and any NOTIFY on that channel triggers your linked agents.

PostgreSQL trigger example
CREATE OR REPLACE FUNCTION notify_new_order()
RETURNS TRIGGER AS $$
BEGIN
  PERFORM pg_notify(
    'new_orders',
    json_build_object(
      'order_id', NEW.id,
      'customer_id', NEW.customer_id,
      'total', NEW.total
    )::text
  );
  RETURN NEW;
END;
$$ LANGUAGE plpgsql;

The connector maintains a persistent connection and automatically reconnects if dropped.

S3 Connector

Trigger agents when files are uploaded to S3. Configure via SNS or EventBridge, and your agent receives file metadata:

Agent payload
{
  "bucket": "my-bucket",
  "key": "uploads/invoice-2024-001.pdf",
  "size": 245789,
  "event_name": "ObjectCreated:Put",
  "content": { ... }  // Optional: file content
}

Use prefix and suffix filters to only process specific paths or file types. Enable include_content to download and pass file contents directly to your agent (up to 10MB).

SQS Connector

Consume messages from AWS SQS queues. The connector handles visibility timeouts, automatic message deletion after successful processing, and dead letter queue routing for failed messages.

Connector Logs

Every connector now maintains a log stream accessible from the connector detail page. See:

  • Connection events (connected, disconnected, reconnecting)
  • Incoming messages with timestamps and payload previews
  • Dispatch events showing which agents were triggered
  • Errors with details for debugging

Unified Connector Management

All connectors now share a consistent create/edit UI. Configuration fields adapt to the connector type, with inline validation and clearer error messages.

More from the Blog

Product Spotlight

Agent Approvals: Human-in-the-Loop for Production AI

AI agents that delete data, process refunds, or call external APIs need a safety net. Connic Approvals pause agent execution at critical moments, wait for human review, and resume automatically — giving you control without killing autonomy.

April 5, 202610 min read
Tutorial

Migrate from LangChain to Production AI Agents

Your LangChain prototype works. Now you need it to handle real traffic. Learn how to migrate existing agent code to a production-grade platform without rewriting from scratch.

March 23, 202611 min read
Product Spotlight

Secure AI Agents: A Production Safety Checklist

Shipping AI agents without a security strategy is a liability. A practical checklist covering prompt injection, PII handling, output validation, and the guardrails you need before go-live.

March 21, 202612 min read
Tutorial

Database vs. Knowledge Base: Choosing the Right Storage

Learn when to use Connic's document database for structured CRUD vs. the knowledge base for semantic search. Configuration tips and best practices.

March 4, 202612 min read
Changelog

What We Shipped in January 2026

Custom observability dashboards with drag-and-drop widgets, model pricing for cost tracking, refreshed connector and runs UI, and llms.txt support.

February 10, 20265 min read
Tutorial

AI Agent Knowledge Base: Setup in 10 Minutes

Your AI agent answers beautifully, just not with your company's information. Learn how RAG transforms generic chatbots into domain experts.

November 15, 20256 min read