Labor-market data.
Within reach.¶
Give your AI tools a shared language for jobs, skills, and careers. Connect to Lightcast’s labor-market intelligence through one focused MCP server.
Python 3.14+ · Local or HTTP · Open source
From a question to useful data¶
01 / UNDERSTAND
Make language consistent¶
Search skills, standardize messy titles, and extract the skills explicitly mentioned in a job description.
02 / CONNECT
See how work relates¶
Map occupations, compare skills, and explore career transitions using Lightcast’s published models.
03 / INVESTIGATE
Put demand in context¶
Explore employment, earnings, education, US job postings, and occupation benchmarks with explicit dataset provenance.
A small setup. A useful connection.¶
# With uv installed and your credentials in .env
uvx --python 3.14 mcp-lightcast@0.4.1 --env-file .env --validate-config
uvx --python 3.14 mcp-lightcast@0.4.1 --env-file .env --transport stdio
Add the server to your MCP client, then ask something concrete: “Extract the skills in this job description and find related skills.”
Built around the current API contracts.
Scope-specific authentication, live version discovery, bounded retries, and honest partial results. The November 2026 review documents the changes, the official sources, and exactly what was tested.