mcp-agent vs declarative-agents
MCP-connected agents compared with YAML-defined workflows
mcp-agent and declarative-agents both help developers build tool-augmented LLM agents. mcp-agent focuses on MCP integrations and composable workflows in Python, while declarative-agents uses a Go runtime to interpret YAML-defined agent behavior.
mcp-agent: Build Agents with Model Context Protocol
A Python framework for building LLM agents around Model Context Protocol servers and composable workflow patterns. It suits developers who need MCP integrations, multi-agent workflows, or durable execution with Temporal.
declarative-agents: Define LLM Agents with YAML Profiles
Nokia Bell Labs' declarative-agents uses YAML profiles and a Go runtime to define and execute tool-augmented LLM agents. It suits teams that want to change agent workflows through configuration rather than rebuild a binary.
| mcp-agent | declarative-agents | |
|---|---|---|
| Language | Python | Go |
| License | Apache-2.0 | BSD-3-Clause |
| Stars | 8.6k | 5 |
| Forks | 893 | 2 |
| Last analyzed | Oct 3, 2026 | Oct 24, 2026 |
Key differences
- mcp-agent manages connections to MCP servers and offers workflow patterns such as routers and orchestrator-workers; declarative-agents defines tools, states, transitions, signals, and budgets in YAML profiles.
- mcp-agent supports asyncio and optional Temporal execution for durable workflows; declarative-agents centers on a profile-interpreting Go runtime.
- mcp-agent is written in Python and licensed under Apache-2.0; declarative-agents is written in Go and licensed under BSD-3-Clause.
- mcp-agent includes a CLI for scaffolding and deployment, with cloud deployment described as beta; declarative-agents provides configuration validation and build, audit, test, and integration commands through Mage.
- mcp-agent has 8.6k stars and 893 forks and is not archived; declarative-agents has 4 stars and 1 fork, reports one contributor, and is in bug-fix-only maintenance mode.
Choose mcp-agent if you…
- need MCP server integrations and managed connection lifecycles for Python agents.
- want reusable workflow patterns or optional Temporal-backed pause, resume, and retry behavior.
- plan to expose an application or agent as an MCP server.
Choose declarative-agents if you…
- want to change agent workflows through YAML profiles interpreted by a Go runtime.
- need reusable agent and tool blocks, configuration validation, and runnable examples.
- are exploring declarative agent design through its DSL specification and design-pattern paper.
Compare the numbers side by side, or add more projects
This comparison is generated with AI from the OSRepos analyses of both projects. Always check each project's repository and documentation before choosing.