Coral Protocol has released Coral v1, an agent stack designed to standardize the discovery, composition, and operation of AI agents built with different frameworks. The release introduces a runtime based on the Model Context Protocol (MCP), developer tools for orchestration, and a public registry for agent discovery.
The v1 release includes several core components. Coral Server is an MCP-native runtime that facilitates structured agent-to-agent (A2A) communication through threaded, mention-addressed messaging. This system is intended to replace methods like context splicing. The release also provides the Coral CLI and Studio, a set of tools enabling developers to manage local and remote agents, connect them within shared threads, and inspect message telemetry for debugging and performance analysis. A public registry serves as a discovery layer for finding and integrating available agents, allowing developers to publish their creations for others to use.
Coral v1 addresses the lack of a common operational protocol between different agent frameworks, such as LangChain and CrewAI, which can hinder composition. By implementing MCP’s common transport and addressing scheme, the protocol allows specialized agents to coordinate without requiring custom glue code or prompt concatenation. The system uses persistent threads and mention-based targeting to organize agent collaboration and reduce operational overhead.
To demonstrate its architecture, Coral has provided an open-source reference implementation named Anemoi. This implementation utilizes a semi-centralized pattern, featuring a light planner agent that coordinates with specialized worker agents communicating directly over Coral MCP threads. The documented coordination loop follows a plan, execute, critique, and refine cycle.
Anemoi was evaluated on the GAIA benchmark, where it achieved a 52.73% pass@3 score. The test used GPT-4.1-mini as the planner and GPT-4o for the worker agents. Coral reported that this performance surpassed a reproduced OWL setup, which scored 43.63% using identical large language models and tooling. These results provide benchmark-supported evidence that structured A2A coordination can outperform simple prompt chaining, particularly when planner capacity is limited. The Anemoi design also aims to reduce redundant token usage and improve cost efficiency for long-horizon tasks.
Coral Protocol has outlined plans for a usage-based marketplace where developers can list agents with pricing metadata and receive payment per call. The platform allows users to rent agents on demand. However, key monetization features are not yet generally available. The company’s developer page explicitly labels “Pay Per Usage / Get Paid Automatically” and “Hosted checkout” as “coming soon.” Similarly, the plan to offer pay-per-usage payouts on Solana is also designated as a future feature. Teams are advised not to assume the general availability of these payment functions until Coral provides an official update.
