Fetch Metrics

Fetches live GitHub GraphQL metrics — stars, commits, followers, repos — into a structured JSON file.

Module

scripts/fetch_metrics.py

Shared HTTP helpers live in scripts/_github_http.py (_headers, _graphql).

What it fetches

Metric	Source
`total_commits`	GraphQL `contributionCalendar`
`stars`	Sum across all repos
`contributions_last_year`	GitHub contributions API
`followers` / `following`	User profile
`forks`	Sum across all repos
`network_count`	Repo network count
`open_issues_count`	All open issues
`orgs_count`	Organizations membership
`watchers`	Sum across all repos
`repos`	Per-repo: name, stars, age_months, language
`contributions_monthly`	Monthly contribution breakdown

CLI

uv run python -m scripts.fetch_metrics \
  --owner wyattowalsh \
  --repo wyattowalsh \
  --output /tmp/metrics.json

Authentication

Requires GITHUB_TOKEN in the environment (PAT with read:user, public_repo):

export GITHUB_TOKEN=ghp_...
uv run python -m scripts.fetch_metrics --owner wyattowalsh --repo wyattowalsh --output metrics.json

Without a token, public API endpoints are used where possible (lower rate limits apply).

Output shape

{
  "total_commits": 5000,
  "stars": 150,
  "contributions_last_year": 800,
  "followers": 200,
  "forks": 40,
  "network_count": 100,
  "open_issues_count": 7,
  "orgs_count": 2,
  "watchers": 20,
  "repos": [
    { "name": "wyattowalsh", "stars": 50, "age_months": 36, "language": "Python" }
  ],
  "contributions_monthly": {
    "2026-01": 120,
    "2026-02": 95
  }
}

Caching

Results are cached in logs/json/YYYY-MM-DD.json (excluded from git via .gitignore). On the same day, the cached file is returned without hitting the API.

Generative Art

Uses metrics as the entropy source for art generation

Ink Garden

Procedural garden seeded from metrics data

Fetch History

Companion module for commit history collection