About dir-mcp
dir-mcp connects AI assistants and IDEs to the AGNTCY Agent Directory for working with OASF agent records. It supports validation, discovery, publishing, signature checks, and format conversion through the dirctl CLI.
Key features
- 01Validate OASF agent records
- 02Search, publish, and retrieve records
- 03Verify signatures and domain ownership
- 04Import and export multiple formats
- 05Guided MCP prompt workflows
Use cases
- 01Validate agent metadata before publishing
- 02Find agent records by directory search
- 03Convert records between MCP, A2A, and other formats
MCP support
The MCP server exposes tools and guided prompts for managing OASF records in Directory infrastructure. It runs through dirctl and can start a local Directory daemon when configured for a local server.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project READMEMCP Server for Directory
Model Context Protocol (MCP) server for working with OASF agent records in the AGNTCY Agent Directory.
The server runs via the dirctl CLI and connects AI assistants and IDEs to Directory infrastructure. Use it to validate and publish agent metadata, search and discover records, navigate OASF skill and domain taxonomies, and generate records from a codebase.
Capabilities:
- Work with OASF schemas and validate agent records
- Search, push, and pull records from Directory servers
- Verify record signatures and domain name ownership
- Import from and export to MCP, A2A, Agent Skills, and GitHub Copilot formats
- Start an embedded local Directory daemon automatically when the configured server is local (disable with
DIR_MCP_DAEMON=false) - Run guided workflows via MCP prompts (create, validate, search, and more)
See docs/directory-mcp.md for full documentation — tools, prompts, setup, configuration, and usage.
Contributing
Contributions are what make the open source community such an amazing place to learn, inspire, and create. Any contributions you make are greatly appreciated. For detailed contributing guidelines, please see CONTRIBUTING.md.
License
Distributed under the Apache-2.0 License. See LICENSE.md for more information.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.