GitHub

Developer sourcing via the GitHub API: search users/repos/code, read profiles + repos (languages, stars, activity), list contributors, and run GraphQL for rich candidate dossiers. The API is free (token; ~5k REST + 5k GraphQL/hr, 30 searches/min).

Output rule

When GitHub is the source, always include each person's GitHub profile URL (https://github.com/<login>) — the primary handle. Pair it with their website + all socials (not just LinkedIn) and an impact tier from their top repo's stars.

One GraphQL dossier per dev

Richest call — captures what REST misses:

user(login:$login){ name websiteUrl followers{totalCount}
  socialAccounts(first:10){nodes{provider url}}
  repositories(first:5, ownerAffiliations:[OWNER,ORGANIZATION_MEMBER,COLLABORATOR],
               orderBy:{field:STARGAZERS,direction:DESC}){nodes{nameWithOwner stargazerCount}} } }

Impact tier

Proxy for the un-exposed achievement badges — top-repo stars:

• ≥5,000 = viral · ≥1,000 = strong · ≥100 = notable

Context-adjust: stars in a niche/recent repo weigh more than in JS; true "going viral" is star velocity (trending), readable from starred_at.

Email harvest

find_user_emails harvests commit author emails (works even when the profile email is hidden). It returns both personal and employer emails.

hyreflow tools execute github search_users \
  --payload '{"q":"language:Go location:Berlin","per_page":10}'

Notes: the GraphQL response is returned unwrapped — the top-level key is user (not data.user). websiteUrl is often the richest contact hub. Best for code-writing roles; skip GitHub for infra/sysadmin roles — see IT Sourcing.