Prefect
Workflow orchestration, now stewarding FastMCP.
Prefect builds dataflow and workflow orchestration infrastructure, and in 2025 took stewardship of FastMCP — the most popular way to author MCP servers. A data-platform company moving squarely into the agent harness layer.
Open-source project
In the news
Best of Agent Harnesses tracks 167 harnesses — and their graveyard
The Best of Agent Harnesses list ranks 167 agent harnesses weekly by capability, autonomy and recovery, ships an MCP server so agents can pick their own harness, and maintains a graveyard that flagged a 268k-star repo for suspected star manipulation. Curation itself becomes harness infrastructure.
OpenAI open-sources Symphony, its codex orchestration harness
OpenAI releases Symphony — open-source orchestration that turns project work into isolated, autonomous coding-agent implementation runs and surfaces the ones that pass, so teams manage outcomes instead of supervising sessions. Big Tech keeps raising the floor on what counts as a baseline harness.
Anthropic moves MCP toward open foundation governance
Anthropic announces plans to donate MCP to a neutral foundation, with new registry, auth and sandbox specs — cementing MCP as vendor-neutral infrastructure for the entire agent ecosystem.
Anthropic introduces the Model Context Protocol
Anthropic open-sources MCP, a standard for connecting AI assistants to data and tools. Within a year it was adopted by OpenAI, Google, Microsoft and most of the agent ecosystem — becoming the de-facto tool-connect layer of agent harnesses.
Related startups
Modal
Serverless cloud built for AI and agentic compute.
Modal lets teams run inference, fine-tuning and arbitrary agent workloads as serverless Python without managing infrastructure — a common substrate for harness backends at scale.
Temporal
Durable execution for long-running agent workflows.
Temporal's durable execution engine is widely adopted to make agent pipelines reliable: state survives crashes, retries are automatic, and workflows run for minutes or months. A key piece of serious harness infrastructure.
Inngest
Durable functions and workflows for AI agents.
Inngest runs agent logic as durable steps — queueing, retries, concurrency control and human-in-the-loop waits — without operating your own workflow infrastructure.