🏁 Script executed:

cat .github/workflows/docs-preview-local.yml | wc -l

Repository: elastic/docs-builder

Length of output: 68

🏁 Script executed:

sed -n '420,460p' .github/workflows/docs-preview-local.yml

Repository: elastic/docs-builder

Length of output: 1297

🏁 Script executed:

# Search for context around the comment upsert logic
sed -n '400,480p' .github/workflows/docs-preview-local.yml

Repository: elastic/docs-builder

Length of output: 2169

🌐 Web query:

GitHub Actions @actions/github paginate listComments API documentation

💡 Result:

@actions/github “paginate” for listComments: use Octokit pagination (Link-header driven) on the REST endpoint that lists issue comments (issues/{issue_number}/comments). The REST endpoint supports pagination via the page and per_page query parameters (max 100) and also supports since filtering; GitHub paginates results using the response Link header when there are more results. [1][2][3] In GitHub Actions, the @actions/github package provides a hydrated Octokit client (github.getOctokit / octokit.rest.*). [4] Use octokit.paginate with the list-comments endpoint so it automatically fetches all pages until the last page, returning a single array of items. [2][3] Example (actions/github-script / Node JS): const comments = await github.paginate(github.rest.issues.listComments, { owner: context.repo.owner, repo: context.repo.repo, issue_number: 123, per_page: 100, }); This pattern is also used in the official actions/github-script discussion as the fix when listComments exceeds the single-page default of 30. [5] Key REST details for listComments / issue comments: - Endpoint: GET /repos/{owner}/{repo}/issues/{issue_number}/comments [1] - Pagination query parameters: page (default 1), per_page (default 30, max 100) [1] Octokit pagination mechanics: - GitHub Docs’ pagination guide states that Octokit’s paginate fetches next pages until the last page and returns all items as a single array. [3]

Citations:

• 1: https://docs.github.com/v3/issues/comments
• 2: https://docs.github.com/rest/using-the-rest-api/using-pagination-in-the-rest-api
• 3: https://docs.github.com/en/rest/using-the-rest-api/using-pagination-in-the-rest-api?apiVersion=2026-03-10
• 4: https://github.com/actions/toolkit/tree/main/packages/github
• 5: How to retrieve all PR/issue comments? actions/github-script#474

🌐 Web query:

@actions/github github.paginate method documentation

💡 Result:

In GitHub Actions, @actions/github exposes an authenticated Octokit client and supports pagination via octokit.paginate (and related iterator helpers) when using the underlying Octokit REST pagination plugin. What documentation exists for github.paginate / octokit.paginate 1) Official GitHub REST API pagination docs (Octokit integration) GitHub documents that for paginated REST API results you can use octokit.paginate, which fetches the next page until the last page and returns all items as a single array; it also supports a map function and octokit.paginate.iterator to iterate page-by-page. [1][2] 2) Authoritative Octokit plugin documentation: @octokit/plugin-paginate-rest The plugin-paginate-rest package documentation states that it adds octokit.paginate which accepts the same parameters as octokit.request; it wraps octokit.request and continues requesting while there is a rel="next" link in the response Link header. It also notes only “List...” endpoints are supported for pagination and that non-list endpoints behave like octokit.request. [3] What the method does (high-confidence behavior) - octokit.paginate wraps octokit.request and keeps requesting the “next” URL as long as the response Link header contains rel="next". [3] - It returns all fetched items as a single array (even when the underlying raw REST result was an object, octokit.paginate returns an array of items). [1][2] - You can supply a map function to process each page’s results and optionally end early by calling done. [1] - Alternatively, octokit.paginate.iterator lets you iterate one page/response at a time, more memory-efficient than collecting everything. [1] How this maps to @actions/github - @actions/github is a hydrated Octokit client returned by getOctokit(token). [4][5] - As part of the standard usage, the client retains pagination APIs (github.paginate) consistent with Octokit.js (used in github-script examples). For example, actions/github-script shows calling await github.paginate(opts) to get all pages of a REST list endpoint. [6] Practical example pattern (from official docs) - For REST: use octokit.paginate with the REST “list” endpoint route/params, and optionally map. GitHub’s docs explicitly recommend octokit.paginate for fetching paginated results via Octokit.js. [1] If you meant pagination for GraphQL - There is a separate plugin for GraphQL pagination: octokit.graphql.paginate, which paginates by scanning for pageInfo and using endCursor when hasNextPage is true. [7] Most relevant references to read - GitHub Docs: “Using pagination in the REST API” (Octokit.paginate + iterator + map function). [1][2] - Octokit plugin: @octokit/plugin-paginate-rest.js (how octokit.paginate is implemented via Link header rel="next"). [3] - @actions/github package overview (shows it returns an authenticated Octokit client and is how Actions code gets an Octokit instance). [4][5]

Citations:

• 1: https://docs.github.com/rest/using-the-rest-api/using-pagination-in-the-rest-api
• 2: https://docs.github.com/en/rest/using-the-rest-api/using-pagination-in-the-rest-api?apiVersion=2026-03-10
• 3: https://github.com/octokit/plugin-paginate-rest.js/
• 4: https://www.npmjs.com/package/@actions/github
• 5: https://github.com/actions/toolkit/tree/main/packages/github
• 6: https://github.com/actions/github-script/
• 7: https://github.com/octokit/plugin-paginate-graphql.js/