Nibleaf public OpenAPI
A typed OpenAPI 3.1 contract for reading published sites, pages, search results, changelogs, sitemaps, and agent indexes. Every operation has a unique operationId and description.
Open openapi.jsonDiscover the supported public reader API, machine-readable product files, Markdown negotiation contract, and command-line tooling without guessing private dashboard routes.
A typed OpenAPI 3.1 contract for reading published sites, pages, search results, changelogs, sitemaps, and agent indexes. Every operation has a unique operationId and description.
Open openapi.jsonStart with llms.txt, request canonical marketing URLs with Accept: text/markdown, and use each published site’s own sitemap and llms files for recovery and retrieval.
Read llms.txtSource main includes project-bound MCP resources and tools for content, search diagnostics, usage, add-ons, integrations, operations, and themes. Scoped keys can expire; write operations are not exposed.
Read the MCP guideThe @nibleaf/cli package inspects a Nibleaf site’s machine-readable endpoints and fetches canonical pages as Markdown for scripts and agents.
View the npm packagenpx @nibleaf/cli inspect https://nibleaf.com
npx @nibleaf/cli fetch https://nibleaf.com/developersThe CLI uses standard HTTP endpoints and does not collect credentials. Its inspect command checks status, content type, OpenAPI, llms, sitemap, Markdown negotiation, and real 404 behavior.
Public published sites require no credentials. Private reader sites use their configured reader session and intentionally return 404 to unauthorized callers. Dashboard endpoints use an authenticated browser session and are not a supported third-party write API. Do not send session cookies or secrets to the CLI. The MCP contract in source main is a separate project-bound, scoped, read-only surface; it is not a dashboard-session write API.