From 28ed3cebb33d7e628c3b01a6c05f090bb74b6a92 Mon Sep 17 00:00:00 2001 From: "fern-api[bot]" <115122769+fern-api[bot]@users.noreply.github.com> Date: Thu, 23 Jul 2026 16:19:50 +0000 Subject: [PATCH] docs: note Vale linting of llms-ignore blocks and cross-link Vale and LLM output pages --- fern/products/docs/pages/ai/llms-txt/customize-llms-txt.mdx | 2 ++ fern/products/docs/pages/developer-tools/vale.mdx | 2 ++ 2 files changed, 4 insertions(+) diff --git a/fern/products/docs/pages/ai/llms-txt/customize-llms-txt.mdx b/fern/products/docs/pages/ai/llms-txt/customize-llms-txt.mdx index 48d1b3c5b6..a22b4eeee2 100644 --- a/fern/products/docs/pages/ai/llms-txt/customize-llms-txt.mdx +++ b/fern/products/docs/pages/ai/llms-txt/customize-llms-txt.mdx @@ -81,6 +81,8 @@ The guiding principle: **UI-only elements belong to human readers, and their pro To preview and debug what AI sees for any page, you can append `.md` or `.mdx` to the page URL to [view its Markdown source](/learn/docs/ai-features/markdown). +Both blocks stay in your MDX source, so source-level tooling like [Vale](/learn/docs/developer-tools/vale) lints their contents regardless of which audience receives the rendered output. + ## Filter endpoint output Filter `llms.txt` output with the `lang` and `excludeSpec` query parameters to reduce token usage. Parameters can also be combined. diff --git a/fern/products/docs/pages/developer-tools/vale.mdx b/fern/products/docs/pages/developer-tools/vale.mdx index 644967e924..f43326b84a 100644 --- a/fern/products/docs/pages/developer-tools/vale.mdx +++ b/fern/products/docs/pages/developer-tools/vale.mdx @@ -64,6 +64,8 @@ const plant = await client.createPlant({ Vale will start checking this text again. ```` +Vale lints your MDX source, so content inside [`` and ``](/learn/docs/ai-features/customize-llm-output#filter-within-a-page) blocks is still checked even though those tags filter what humans and LLMs receive. Wrap a block in `vale off`/`vale on` to exclude it from linting. + Consider integrating Vale into your workflow so it runs automatically for all contributors: