Agileday

You deploy an MCP server that connects Agileday’s skills and employee data with large language models, enabling natural language queries about your organization’s competencies, experts, and competence profiles. It’s useful for surfacing talent, mapping skills to projects, and accelerating answers from your AI assistant.

How to use

You connect your MCP client (for example Claude Desktop) to the Agileday MCP server, then ask natural language questions such as who knows a skill, who are the experts in a given domain, or what is someone’s competence profile. The server exposes three core tools you can rely on: finding experts with a skill, retrieving an employee’s competence profile, and listing the organization’s skills in a categorized taxonomy.

Practical usage patterns include:

• Ask for experts with a specific skill to assemble a quick talent map.
• Retrieve a person’s top skills and motivation to tailor opportunities.
• Explore the full skill taxonomy to understand terminology and guide AI queries.

How to install

Prerequisites you need before installing:

• Docker (recommended) or Python for local development
• An Agileday tenant ID and API token for access to data
• An MCP client that can load and run MCP servers (for example Claude Desktop) and point it at the MCP URL or stdio server.

# Option A: Run with Claude Desktop using Docker (stdio)
# This starts a container that serves the MCP via standard input/output.
docker run -i --rm \
  -e AGILEDAY_TENANT_ID=your_tenant_id \
  -e AGILEDAY_API_TOKEN=your_api_token \
  ghcr.io/eficode/mcp-agileday:latest

# Option B: Run via Docker with a prebuilt image (stdio)
docker run -i --rm \
  -e AGILEDAY_TENANT_ID=your_tenant_id \
  -e AGILEDAY_API_TOKEN=your_api_token \
  agileday-mcp-server

# Option C: Run locally with Python (stdio)
git clone https://github.com/yourusername/agileday-mcp-server.git
cd agileday-mcp-server
pip install -r requirements.txt
python3 /absolute/path/to/agileday_server.py

# Option D: Run as a standalone HTTP service (http transport)
docker run -d \
  -p 8000:8000 \
  -e AGILEDAY_TENANT_ID=your_tenant \
  -e AGILEDAY_API_TOKEN=your_token \
  ghcr.io/eficode/mcp-agileday:latest \
  --transport http

Additional configuration notes

Environment variables you need to set in all options are:

• AGILEDAY_TENANT_ID: Your company slug (subdomain). For example, acme to use acme.agileday.io
• AGILEDAY_API_TOKEN: Your API Bearer token These values are required to access Agileday data.

The server exposes a universal endpoint for HTTP transport and supports a health check endpoint when running in HTTP mode. If you run with HTTP transport, you can reach the MCP at the configured host and port.

Security and best practices

Do not commit your API token to version control. Pass it only through environment variables or your local secure config. Use a read-only API token if possible, since the MCP client only fetches data.

Troubleshooting tips

If you cannot connect from your MCP client:

• Verify AGILEDAY_TENANT_ID and AGILEDAY_API_TOKEN are correct.
• Check that the container is running and accessible on the expected port when using HTTP transport.
• Ensure your client is configured to use the correct MCP server name and endpoint.

Notes

The server provides three capabilities: find_experts_with_skill, get_employee_competence_profile, and list_organization_skills. Use these tools to search for experts, retrieve profiles, and understand the organization’s skill taxonomy.

Available tools

find_experts_with_skill

Query for employees who possess a specific skill and return their proficiency levels.

get_employee_competence_profile

Fetch a full list of skills, proficiency levels, and motivation for a specific employee.

list_organization_skills

Retrieve the full taxonomy of skills organized by category for the workspace.