agentleFS
ElasticKnown publisherelastic/docs-builder/src/Elastic.Markdown/AGENTS.md

docs-builder / Elastic.Markdown

AGENTS.md25 starsChanged 19 days agoApache-2.0View on GitHub

The core Markdown parser: a Markdig extension that adds Myst-style directives and roles, plus the diagnostics, linting, and export machinery that turns a directory of .md files into rendered pages. Everything else in this repo (the CLI, the frontend, essc) consumes this project's output. Most directives are a 3-file folder under Myst/Directives/<Name>/: <Name>Block.cs (the Markdig block type), <Name>View.cshtml (Razor rendering), <Name>ViewModel.cs (data the view reads). Some have more files (e.g. Tabs/, Stepper/, Image/ render multiple sub-elements), and at least one…

We split this AGENTS.md into the rules it gives an agent

1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Document the syntax in docs/syntax/directives.md.

    line 59

≈ “Common” is an estimate: at least 1 in 5 of the 36,368 AGENTS.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
AGENTS.md

Elastic.Markdown

The core Markdown parser: a Markdig extension that adds Myst-style directives and roles, plus the diagnostics, linting, and export machinery that turns a directory of .md files into rendered pages. Everything else in this repo (the CLI, the frontend, essc) consumes this project's output.

Structure

Myst/
  Directives/            Block-level directives ({admonition}, {tabs}, {dropdown}, ...). One
                          folder per directive, typically Block.cs + View.cshtml + ViewModel.cs.
  Roles/                 Inline analog of directives ({applies_to}, {kbd}, {math}, icons). Same
                          idea, much smaller surface (Role.cs, RoleParser.cs, one folder per role).
  FrontMatter/            YAML front matter parsing.
  AppliesTo/              The applies-to versioning/compatibility model shared by directives+roles.
  CodeBlocks/             Fenced code block handling (substitutions, callouts, console/copy UI).
  InlineParsers/          Custom inline Markdig parsers (substitutions, cross-doc links, etc).
  Linters/                Structural lint rules over the parsed document tree.
  Components/             Shared Myst-level building blocks used by multiple directives/roles.
  Renderers/              HTML rendering glue for the extension.
Diagnostics/              Build-time diagnostics (broken links, missing includes, etc) — separate
                          from Myst/Linters, which is about document-structure rules.
Exporters/                Turn a built doc set into something else: plain text, LLM-friendly
                          markdown, or Elasticsearch documents (Exporters/Elasticsearch/ —
                          feeds the shared Elastic.Documentation.Search.Contract schema).
Extensions/               Cross-cutting Markdown extensions (CLI reference, detection rules).
IO/                       File system abstractions.
Page/                     Assembled-page model consumed by the CLI/site.
Components/               Design-system slices for Markdown output (admonition, tabs, stepper, code block, image, ...). Their CSS lives
                          in the site bundle under Elastic.Documentation.Site/Components/Markdown/.
Layout/                   Shared Razor layout partials.

Directives: the pattern, and where it bends

Most directives are a 3-file folder under Myst/Directives/<Name>/: <Name>Block.cs (the Markdig block type), <Name>View.cshtml (Razor rendering), <Name>ViewModel.cs (data the view reads). Some have more files (e.g. Tabs/, Stepper/, Image/ render multiple sub-elements), and at least one composes rather than duplicates: DropdownBlock is a one-line subclass of AdmonitionBlock (Admonition/AdmonitionBlock.cs) rather than its own implementation — check whether a new directive is really a variant of an existing one before writing a fresh block type.

Dispatch is a plain if chain, not a lookup table. DirectiveBlockParser.CreateFencedBlock (Myst/Directives/DirectiveBlockParser.cs) matches the directive name string via a sequence of if (info.IndexOf("{name}") > 0) return new XBlock(...) checks. The FrozenDictionary in that same file (UnsupportedBlocks) is only the legacy/removed-directive tracking list, not the active registration mechanism — don't assume adding a directive means adding to a dictionary.

Adding a new directive: the full cross-project checklist

A new directive ripples outside this project. In order:

  1. Myst/Directives/<Name>/ — <Name>Block.cs + <Name>View.cshtml + <Name>ViewModel.cs.
  2. Wire it into DirectiveBlockParser.CreateFencedBlock's if chain.
  3. src/Elastic.Documentation.Site/Assets/markdown/<name>.css — directive CSS lives in the frontend project, named after the directive (see admonition.css, dropdown.css, button.css there).
  4. Optional interactive behavior → TS in src/Elastic.Documentation.Site/Assets/.
  5. Document the syntax in docs/syntax/directives.md.

Missing step 3 or 5 is the most common way a directive works but looks broken or undocumented.

Testing

tests/Elastic.Markdown.Tests/ — base classes in Inline/InlneBaseTests.cs (InlineTest, LeafTest<T>, BlockTest<T>, InlineTest<T>) parse a raw markdown string against a MockFileSystem fixture built by TestHelpers.CreateConfigurationContext(...) and expose Html/ Document/Collector for assertions. Subclass the matching base, pass markdown via a primary constructor, assert with AwesomeAssertions (Html.Should().Contain(...)). Run with dotnet test tests/Elastic.Markdown.Tests/.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.