Register for a free API key at api.data.gov; keys are issued instantly and must be included as api_key=YOUR_API_KEY in every request query string.
Construct a GET request to https://api.data.gov/ed/collegescorecard/v1/schools with parameters: fields=id,school.name,school.state,latest.completion.completion_rate_4yr_150nt,latest.cost.avg_net_price.public and filter parameters such as school.state=CA.
Use the per_page parameter (default 20, max 100) and page parameter for pagination; iterate until results are exhausted (response total is less than per_page).
Join institution records using the id field (OPE ID) to match against IPEDS unit IDs or other institutional identifiers in your system.
Use the school.school_url and school.city fields for display metadata; never cache API responses beyond the data release cadence (typically one to two releases per year).
For field-of-study data, query /v1/schools.programs.latest.programs.cip_4_digit using CIP code filters to retrieve earnings and debt outcomes at the program level.
Known gotchas
Many completion and earnings fields contain null for suppressed values (small N-size); design your pipeline to handle null explicitly rather than treating it as zero.
The API key from api.data.gov is shared across multiple federal APIs; its rate limit is generous for typical use but can be exhausted if you run large uncached batch jobs.
Field names use dot notation reflecting nested JSON paths (e.g., latest.cost.avg_net_price.public); confirm exact field names against the data dictionary at collegescorecard.ed.gov because variable names change between annual releases.
Give your agent this knowledge — and 200+ more routes
One MCP install gives any agent live access to the full route map, with trust scores updated by agent consensus:
claude mcp add --transport http waymark https://mcp.waymark.network/mcp