diff --git a/CHANGELOG.md b/CHANGELOG.md
index e9138e2..495c3e9 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -6,6 +6,7 @@
 - Helpers
   - `Ema.Helpers.PathTree` moved to separate package *pathtree*.
   - `Ema.Helpers.FileSystem` moved to separate package *unionmount*.
+  - `Ema.Helpers.Markdown` moved to separate package *commonmark-simple*.
 
 ## 0.4.0.0 -- 2022-01-19
 
diff --git a/docs/guide/helpers.md b/docs/guide/helpers.md
index b9da40c..67237d4 100644
--- a/docs/guide/helpers.md
+++ b/docs/guide/helpers.md
@@ -1,10 +1,10 @@
 ---
 order: 5
 ---
-# Helpers
+# Howto
 
-Beyond the model and route types, Ema leaves it up to you as to how to develop your site. The following are not *required* when using Ema; nevertheless they are useful inasmuch as they capture common patterns in writing a good static site:
+Beyond the model and route types, Ema leaves it up to you as to how to develop your site. Here are some common aspects of writing a static site:
 
-* [Blaze](guide/helpers/blaze.md) -- [We recommend--but not mandate--Tailwind for CSS and blaze-html as HTML DSL]{.item-intro}
+* [Blaze HTML DSL](guide/helpers/blaze.md) -- [Use blaze-html with Tailwind CSS]{.item-intro}
 * [Working with files](guide/helpers/filesystem.md) -- [Use `unionmount` to support hot-reload on files]{.item-intro}
 * [Converting Markdown](guide/helpers/markdown.md) -- [Pointers on how to work with Markdown files]{.item-intro}
\ No newline at end of file
diff --git a/docs/guide/helpers/markdown.md b/docs/guide/helpers/markdown.md
index 7d764cc..c42a1ac 100644
--- a/docs/guide/helpers/markdown.md
+++ b/docs/guide/helpers/markdown.md
@@ -8,14 +8,15 @@ There are quite a few packages to convert Markdown to HTML,
 
 - [Pandoc](https://hackage.haskell.org/package/pandoc) -- [Supports formats other than Markdown]{.item-intro}
 - [commonmark-hs](https://github.com/jgm/commonmark-hs) -- [Lightweight parser by the same author of Pandoc]{.item-intro}
+    - [commonmark-simple](https://hackage.haskell.org/package/commonmark-simple-0.1.0.0) -- [Simpler interface to the above, with frontmatter support]{.item-intro}
 - [mmark](https://github.com/mmark-md/mmark) -- [*Strict* Markdown parser]{.item-intro}
 
-## Helper
+## `commonmark-simple`
 
-Ema provides a helper to parse Markdown files with YAML frontmatter, using commonmark-hs. If you are parsing front matter, you can use any type that has a [`FromYAML`](https://hackage.haskell.org/package/HsYAML-0.2.1.0/docs/Data-YAML.html#t:FromYAML) instance.
+`commonmark-simple` uses commonmark-hs to provide a simpler API, along with front matter support. If you are parsing front matter, you can use any type that has a [`FromYAML`](https://hackage.haskell.org/package/HsYAML-0.2.1.0/docs/Data-YAML.html#t:FromYAML) instance.
 
 ```haskell
-import qualified Ema.Helper.Markdown as Markdown
+import qualified Commonmark.Simple as CS
 
 -- Front matter metadata can be any type with a `FromYAML` instance
 -- 
@@ -24,10 +25,10 @@ import qualified Ema.Helper.Markdown as Markdown
 type Metadata = Map Text Text 
 
 -- Returns `Either Text (Metadata, Pandoc)`
-Markdown.parseMarkdownWithFrontMatter @Metadata 
+CS.parseMarkdownWithFrontMatter @Metadata 
     "test.md" "Hello *world*"
 ```
 
 The template repo, as well as [Emanote](https://github.com/srid/emanote) (used to generate this site), uses this helper to parse Markdown files into Pandoc AST. Consult [the template repo's source code](https://github.com/srid/ema-template/blob/master/src/Main.hs) for details.
 
-Note that with Ema you can get [hot reload](concepts/hot-reload.md) support for your Markdown files using [filesystem notifications](guide/helpers/filesystem.md).
+Note that with Ema you can get [hot reload](concepts/hot-reload.md) support for your Markdown files using [the `unionmount` package](guide/helpers/filesystem.md).
diff --git a/ema.cabal b/ema.cabal
index 04eeaaf..fde1f08 100644
--- a/ema.cabal
+++ b/ema.cabal
@@ -1,6 +1,6 @@
 cabal-version:      2.4
 name:               ema
-version:            0.5.2.0
+version:            0.5.3.0
 license:            AGPL-3.0-only
 copyright:          2021 Sridhar Ratnakumar
 maintainer:         srid@srid.ca
@@ -68,15 +68,6 @@ library
     build-depends:
       , blaze-html
       , blaze-markup
-      , commonmark
-      , commonmark-extensions
-      , commonmark-pandoc
-      , megaparsec
-      , pandoc-types
-      , parsec
-      , parser-combinators
-      , time
-      , yaml
 
     if flag(with-examples)
       build-depends: time
@@ -139,7 +130,6 @@ library
   if (flag(with-helpers) || flag(with-examples))
     exposed-modules:
       Ema.Helper.Blaze
-      Ema.Helper.Markdown
 
   other-modules:
     Ema.App
diff --git a/src/Ema/Helper/Markdown.hs b/src/Ema/Helper/Markdown.hs
deleted file mode 100644
index a97d3c5..0000000
--- a/src/Ema/Helper/Markdown.hs
+++ /dev/null
@@ -1,132 +0,0 @@
--- | Helper to deal with Markdown files
---
--- TODO: Publish this eventually to Hackage, along with wiki-link stuff from
--- emanote (maybe as separate package).
-module Ema.Helper.Markdown
-  ( -- Parsing
-    -- TODO: Publish to Hackage as commonmark-pandoc-simple?
-    parseMarkdownWithFrontMatter,
-    parseMarkdown,
-    fullMarkdownSpec,
-    -- Utilities
-    plainify,
-  )
-where
-
-import Commonmark qualified as CM
-import Commonmark.Extensions qualified as CE
-import Commonmark.Pandoc qualified as CP
-import Control.Monad.Combinators (manyTill)
-import Data.Aeson (FromJSON)
-import Data.Yaml qualified as Y
-import Text.Megaparsec qualified as M
-import Text.Megaparsec.Char qualified as M
-import Text.Pandoc.Builder qualified as B
-import Text.Pandoc.Definition (Pandoc (..))
-import Text.Pandoc.Walk qualified as W
-
--- | Parse a Markdown file using commonmark-hs with all extensions enabled
-parseMarkdownWithFrontMatter ::
-  forall meta m il bl.
-  ( FromJSON meta,
-    m ~ Either CM.ParseError,
-    bl ~ CP.Cm () B.Blocks,
-    il ~ CP.Cm () B.Inlines
-  ) =>
-  CM.SyntaxSpec m il bl ->
-  -- | Path to file associated with this Markdown
-  FilePath ->
-  -- | Markdown text to parse
-  Text ->
-  Either Text (Maybe meta, Pandoc)
-parseMarkdownWithFrontMatter spec fn s = do
-  (mMeta, markdown) <- partitionMarkdown fn s
-  mMetaVal <- first show $ (Y.decodeEither' . encodeUtf8) `traverse` mMeta
-  blocks <- first show $ join $ CM.commonmarkWith @(Either CM.ParseError) spec fn markdown
-  let doc = Pandoc mempty $ B.toList . CP.unCm @() @B.Blocks $ blocks
-  pure (mMetaVal, doc)
-
-parseMarkdown :: FilePath -> Text -> Either Text Pandoc
-parseMarkdown fn s = do
-  cmBlocks <- first show $ join $ CM.commonmarkWith @(Either CM.ParseError) fullMarkdownSpec fn s
-  let blocks = B.toList . CP.unCm @() @B.Blocks $ cmBlocks
-  pure $ Pandoc mempty blocks
-
-type SyntaxSpec' m il bl =
-  ( Monad m,
-    CM.IsBlock il bl,
-    CM.IsInline il,
-    Typeable m,
-    Typeable il,
-    Typeable bl,
-    CE.HasEmoji il,
-    CE.HasStrikethrough il,
-    CE.HasPipeTable il bl,
-    CE.HasTaskList il bl,
-    CM.ToPlainText il,
-    CE.HasFootnote il bl,
-    CE.HasMath il,
-    CE.HasDefinitionList il bl,
-    CE.HasDiv bl,
-    CE.HasQuoted il,
-    CE.HasSpan il
-  )
-
--- | GFM + official commonmark extensions
-fullMarkdownSpec ::
-  SyntaxSpec' m il bl =>
-  CM.SyntaxSpec m il bl
-fullMarkdownSpec =
-  mconcat
-    [ CE.gfmExtensions,
-      CE.fancyListSpec,
-      CE.footnoteSpec,
-      CE.mathSpec,
-      CE.smartPunctuationSpec,
-      CE.definitionListSpec,
-      CE.attributesSpec,
-      CE.rawAttributeSpec,
-      CE.fencedDivSpec,
-      CE.bracketedSpanSpec,
-      CE.autolinkSpec,
-      CM.defaultSyntaxSpec,
-      -- as the commonmark documentation states, pipeTableSpec should be placed after
-      -- fancyListSpec and defaultSyntaxSpec to avoid bad results when parsing
-      -- non-table lines
-      CE.pipeTableSpec
-    ]
-
--- | Identify metadata block at the top, and split it from markdown body.
---
--- FIXME: https://github.com/srid/neuron/issues/175
-partitionMarkdown :: FilePath -> Text -> Either Text (Maybe Text, Text)
-partitionMarkdown =
-  parse (M.try splitP <|> fmap (Nothing,) M.takeRest)
-  where
-    separatorP :: M.Parsec Void Text ()
-    separatorP =
-      void $ M.string "---" <* M.eol
-    splitP :: M.Parsec Void Text (Maybe Text, Text)
-    splitP = do
-      separatorP
-      a <- toText <$> manyTill M.anySingle (M.try $ M.eol *> separatorP)
-      b <- M.takeRest
-      pure (Just a, b)
-    parse :: M.Parsec Void Text a -> String -> Text -> Either Text a
-    parse p fn s =
-      first (toText . M.errorBundlePretty) $
-        M.parse (p <* M.eof) fn s
-
--- | Convert Pandoc AST inlines to raw text.
-plainify :: [B.Inline] -> Text
-plainify = W.query $ \case
-  B.Str x -> x
-  B.Code _attr x -> x
-  B.Space -> " "
-  B.SoftBreak -> " "
-  B.LineBreak -> " "
-  B.RawInline _fmt s -> s
-  B.Math _mathTyp s -> s
-  -- Ignore the rest of AST nodes, as they are recursively defined in terms of
-  -- `Inline` which `W.query` will traverse again.
-  _ -> ""