geepers-corpus
by lukeslp
Search & ResearchBrief
Query the COCA (Corpus of Contemporary American English) linguistics API for word frequency, collocations.
Description
What This Skill Does
Queries the COCA (Corpus of Contemporary American English) API for word frequency, collocations, concordances, and genre-specific usage data. COCA covers 1+ billion words from spoken, fiction, magazine, newspaper, and academic sources from 1990 to present.
Direct API access to COCA gives programmatic, citation-quality frequency and collocation data that writing tools and dictionaries don't expose.
When to Use It
- Checking how formal a word is before using it in academic writing
- Finding natural verb-noun collocations for non-native English writers
- Comparing word frequency across spoken vs academic genres
- Researching how word usage has shifted over decades
- Verifying whether a phrasing sounds natural in American English
Example Workflow
User asks: What words commonly follow 'run' as a verb in American English?
- 1
Call the collocations endpoint with word=run and pos=verb
- 2
Parse the returned MI scores and frequency counts
- 3
Rank collocates by statistical association strength
- 4
Filter or group by genre if requested
A ranked list of collocates (e.g., 'run risk', 'run course', 'run program') with MI scores and frequency per million words
Requirements
- DREAMER_API_KEY environment variable (from api.dr.eamer.dev)
Installation
clawhub install lukeslp/geepers-corpusRun the CLI command, or paste the prompt into Claude Code, Cursor, Codex or whichever coding agent you use.