diff --git a/.vscode/settings.json b/.vscode/settings.json
index cc4be24..f24f34c 100755
--- a/.vscode/settings.json
+++ b/.vscode/settings.json
@@ -1,7 +1,9 @@
 {
     "cSpell.words": [
+        "autochurch",
         "freevar",
         "pyparsing",
+        "runstatus",
         "subvar"
     ],
     "python.analysis.typeCheckingMode": "basic"
diff --git a/README.md b/README.md
index cd72363..3659894 100644
--- a/README.md
+++ b/README.md
@@ -7,10 +7,8 @@
  - Python files: installable, package list, etc
  - $\alpha$-equivalence check
  - Versioning
- - Automatic church numerals
  - Prettyprint functions (combine args, rename bound variables)
  - Documentation in README
- - Don't crash on errors
 
 ## Todo:
  - live syntax check
@@ -22,4 +20,5 @@
  - Pin header to top of screen
 
 ## Mention in Docs
- - lambda functions only work with single-letter arguments
\ No newline at end of file
+ - lambda functions only work with single-letter arguments
+ - church numerals
\ No newline at end of file
diff --git a/parser.py b/parser.py
index 6e75be1..919edbe 100755
--- a/parser.py
+++ b/parser.py
@@ -1,5 +1,6 @@
 import pyparsing as pp
 import tokens
+import utils
 
 class Parser:
 	"""
@@ -20,6 +21,9 @@ class Parser:
 	pp_macro = pp.Word(pp.alphas + "_")
 	pp_macro.set_parse_action(tokens.macro.from_parse)
 
+	pp_church = pp.Word(pp.nums)
+	pp_church.set_parse_action(utils.autochurch)
+
 	# Function definitions.
 	# Right associative.
 	#
@@ -50,7 +54,7 @@ class Parser:
 	pp_call <<= pp_expr[2, ...]
 	pp_call.set_parse_action(tokens.lambda_apply.from_parse)
 
-	pp_expr <<= pp_lambda_fun ^ (lp + pp_expr + rp) ^ pp_macro ^ (lp + pp_call + rp)
+	pp_expr <<= pp_lambda_fun ^ (lp + pp_expr + rp) ^ pp_macro ^ (lp + pp_call + rp) ^ pp_church
 	pp_all = pp_expr | pp_macro_def
 
 	pp_command = pp.Suppress(":") + pp.Word(pp.alphas + "_")
diff --git a/utils.py b/utils.py
new file mode 100644
index 0000000..8590300
--- /dev/null
+++ b/utils.py
@@ -0,0 +1,24 @@
+import tokens
+
+def autochurch(results):
+	"""
+	Makes a church numeral from an integer.
+	"""
+
+	num = int(results[0])
+
+	f = tokens.bound_variable()
+	a = tokens.bound_variable()
+
+	chain = a
+
+	for i in range(num):
+		chain = tokens.lambda_apply(f, chain)
+
+	return tokens.lambda_func(
+		f,
+		tokens.lambda_func(
+			a,
+			chain
+		)
+	)
\ No newline at end of file