SineFrame M3
AIPytest-based CI gate for MCP servers that tests them against real agents like Claude Code and Codex.
SineFrame M3
Pytest-based CI gate for MCP servers that tests them against real agents like Claude Code and Codex.
About SineFrame M3
SineFrame M3 is an open-source, local-first pytest framework for testing MCP servers. Instead of calling a bare model API, it runs real agent harnesses such as Claude Code, Codex, OpenCode and Pi against your server, records every MCP tool call, and fails a test when the server was skipped or called incorrectly. Direct tests check schemas, responses and errors without a model, and the same suite can run locally and in CI, where a run with zero executed tests fails the gate. Reports and run history are stored in the project, with an optional local viewer.
Screenshots
Key Features
Use Cases
Frequently asked questions about SineFrame M3
What is SineFrame M3?
Pytest-based CI gate for MCP servers that tests them against real agents like Claude Code and Codex.
How does SineFrame M3 work?
SineFrame M3 works by combining Real Agent Harnesses: Runs Claude Code, Codex, OpenCode or Pi binaries so tool choice reflects real discovery and permission flows., Tool Call Assertions: Records every MCP tool call and lets tests assert expected calls and arguments., Direct Server Tests: Checks schemas, results and errors without a model in ordinary pytest files., Cross-Harness Comparison: Runs the same tests across harnesses and versions to catch regressions after agent updates., CI Gate: m3 ci test fails on broken assertions or when no tests executed at all. to help users with MCP Regression Testing: Verify agents still call the right tool after changing tool descriptions., Agent Compatibility: Check whether a server that works in Claude Code also works in Codex or OpenCode., Release Gating: Block pull requests that break MCP behavior in CI., Upgrade Monitoring: Compare behavior across versions of Codex or other harnesses., Debugging Misrouted Calls: Inspect recorded tool calls to see why an agent asked for the wrong argument..
What are the main features of SineFrame M3?
Key features include Real Agent Harnesses: Runs Claude Code, Codex, OpenCode or Pi binaries so tool choice reflects real discovery and permission flows., Tool Call Assertions: Records every MCP tool call and lets tests assert expected calls and arguments., Direct Server Tests: Checks schemas, results and errors without a model in ordinary pytest files., Cross-Harness Comparison: Runs the same tests across harnesses and versions to catch regressions after agent updates., CI Gate: m3 ci test fails on broken assertions or when no tests executed at all..
Who is SineFrame M3 for?
SineFrame M3 is useful for anyone interested in MCP Regression Testing: Verify agents still call the right tool after changing tool descriptions., Agent Compatibility: Check whether a server that works in Claude Code also works in Codex or OpenCode., Release Gating: Block pull requests that break MCP behavior in CI., Upgrade Monitoring: Compare behavior across versions of Codex or other harnesses., Debugging Misrouted Calls: Inspect recorded tool calls to see why an agent asked for the wrong argument..
How much does SineFrame M3 cost?
SineFrame M3 is free to use.
How do I get started with SineFrame M3?
Visit https://m3.sineframe.com/ to sign up and explore SineFrame M3.
Explore more AI Mcps tools
Browse all Mcps tools →Compare SineFrame M3: vs Pomerado · vs Manifold MCP · vs Lune · vs Floot MCP