Airbyte Agent MCP

Airbyte Context Store is a replicated, pre-indexed data layer for AI agents. It ingests curated data from enterprise SaaS sources such as Zendesk, Stripe, HubSpot, Salesforce, and Slack, then indexes records for natural-language search, entity lookup, and cross-source joins so agents can query a unified store instead of repeatedly calling live source APIs. The architecture separates discovery and search in the context store from direct API access for fetching or writing data. Airbyte describes the store as supporting sub-second agent-oriented retrieval and providing the data layer behind interfaces such as its MCP server and Agent CLI.

Airbyte Agent CLI (airbyte-agent) is a Go command-line interface from Airbyte for AI agents, scripts, CI jobs, and shell-capable agent harnesses. It uses JSON input and output, consistent noun-verb commands, a bundled agent skill, and non-interactive operation. Agents can create connectors, inspect entities, actions, and schemas, search and read data, and execute read or write actions across more than 50 business systems through Airbyte's Context Store. The CLI supports schema inspection before operations and reports failures as typed JSON on stderr, with errors such as authentication, not-found, and rate-limit conditions. It can be installed with an install script, Homebrew, or a GitHub release binary, and credentials are kept in a local settings file while third-party credentials are entered through a browser widget.

An MCP server from Zapier that connects authenticated applications so compatible AI tools can access and use them through a single integration layer. It supports clients including Cursor, Claude Code, and Codex.

AgentConnect is an open-source, self-hostable platform that integrates autonomous AI agents (ACP agents such as Claude Code, Codex, and Gemini CLI) into communication platforms including Slack, Telegram, Discord, and GitHub. It provides tagging of agents and facilities to assign roles, tools, and memory so agents can act within those channels.

Apify MCP Server is an MCP server from Apify that lets AI assistants use Apify Actors as tools for web search, data extraction, scraping, browsing, and automation. Its hosted endpoint at mcp.apify.com supports Streamable HTTP and OAuth, while a local stdio distribution is available through the @apify/actors-mcp-server package. The server dynamically discovers Actors from the Apify Store, loads their input schemas, runs them, and exposes results through Apify datasets and key-value stores. It also provides tools for searching Apify documentation, inspecting and managing Actor runs, tasks, schedules, builds, and source files. Tool selection can be restricted through URL parameters or CLI flags, and the server supports agentic payment mechanisms including AGI, direct x402, and Skyfire.