D'AUBE // BRIDGE2 is an open-source skill compiler and MCP bridge. You define one portable skill spec and it generates integration files for several AI agents and MCP hosts.
PluginContact for pricing
Step 1: Clone the repository and run python -m pip install -e ".[dev]", or install the wheel from the GitHub release.
Step 2: Run daube-bridge validate examples/skill.yaml to check a skill spec.
Step 3: Run daube-bridge compile examples/skill.yaml -o build/sample to generate the target files.
Step 4: Start the MCP server with python -m daube_bridge.server for stdio, or daube-bridge serve for the HTTP endpoint at /mcp.
Step 5: Point your MCP client at the server and call bridge_targets, validate_skill or compile_skill.
Integration files for 7 target families, including MCP tool schemas, Claude plugin files, Gemini and OpenAI skill files, function schemas and a Chromium extension manifest.
Yes. Use python -m daube_bridge.server for stdio, or daube-bridge serve for an HTTP endpoint on localhost.
Three: bridge_targets, validate_skill and compile_skill.
MIT.
Requirements
Thalovant MCPThalovant MCP is an MCP server for the Thalovant control-plane and hub runtime APIs. It works over stdio and Streamable HTTP with hosts such as Claude Desktop, Codex and Cursor.
0nMCPUniversal AI API orchestrator exposing a very large multi-service tool catalog.