GitHub Repo Search — Stars, Language & Topics
Pricing
$4.00 / 1,000 results
Search and scrape GitHub repositories by keyword, language, stars, forks, or topic. Extract structured repo metadata including owner, license, topics, and activity timestamps. Sort by stars, forks, or recently updated. Export to JSON, CSV, or API. No token required.
Search query using GitHub syntax. Examples: 'web scraping', 'machine learning language:python', 'topic:nextjs stars:>500'
Pre-configured workflow. Each mode auto-selects enrichments, scoring, and sorting for a specific use case. Leave empty for manual configuration.
How to sort results. Leave unset to let the selected mode pick the best sort (e.g. Trend Watch sorts by recently updated). Defaults to Most Stars when no mode is set.
Filter by programming language (e.g., python, rust, typescript)
Maximum repositories to return. With auto-partition enabled, can exceed 1,000.
Filter out forked repositories from results. Leave unset to let the selected mode decide.
Filter out archived repositories from results. Leave unset to let the selected mode decide.
When a query matches more than 1,000 repos, automatically split into sub-queries by star ranges to retrieve more. Deduplicates across partitions. Leave unset to let the selected mode decide (Market Map enables it).
Compare specific repos side-by-side. Enter full names like 'facebook/react'. Skips search and fetches each repo directly. Scores, ranks, and picks a winner.
Fetch the full profile for each repo: community health, activity, contributors, languages and releases. Asking for everything is the slowest option — if you only need one part, a solution mode requests just that and returns far more repos complete. Check enrichmentSkipped in the output to see what you got.
Extract real email addresses from recent commits. Filters out noreply/bot addresses. Auto-enabled by maintainer-outreach and due-diligence modes.
Detect changes since last run: new repos, score changes, newly abandoned repos. Useful for scheduled monitoring. Stores state in a named key-value store.
Personal access token, no scopes needed. Makes runs faster and more reliable, and means every repo comes back fully enriched instead of some rows returning base data with enrichmentSkipped=true. Free to create at github.com/settings/tokens.
Archive each run's intelligence digest (top repositories, scores, and monitoring changes) to Notion. Create the connector once in Apify Console → Settings → MCP connectors, then select it here. The actor never sees your Notion token.
Target Notion data source to write into. Leave blank to create standalone pages.
summary = one page per run (digest + top repositories). per-repo = one page per top repository.
Post the run's intelligence digest (top repositories and monitoring changes) to your Slack. Create the connector once in Apify Console → Settings → MCP connectors, then select it here. The actor never sees your Slack token.
Channel to post to, e.g. #github-watch. Leave blank to use the connector's default channel.
How many top-ranked repositories to include in the Slack/Notion delivery (1-50). The full set always stays in the dataset.
