diff --git a/.github/workflows/check.yml b/.github/workflows/check.yml
new file mode 100644
index 000000000..cb1610feb
--- /dev/null
+++ b/.github/workflows/check.yml
@@ -0,0 +1,29 @@
+name: "Check formatting and linting"
+
+on:
+  push:
+  pull_request:
+  workflow_dispatch:
+
+permissions:
+  contents: read
+
+jobs:
+  check:
+    name: "just check"
+    runs-on: ubuntu-24.04
+
+    steps:
+      - uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1 # v7.0.1
+        with:
+          persist-credentials: false
+
+      - uses: opensafely-core/setup-action@d9171097dd0d37bdb7bed58f701eb03ff317ed17 # v1.7.0
+        with:
+          python-version: "3.11"
+          install-just: true
+          install-uv: true
+          cache: uv
+
+      - name: Check formatting, linting, and import sorting
+        run: just check
diff --git a/docs/actions-reusable.md b/docs/actions-reusable.md
index 90230a4fc..fc490c833 100644
--- a/docs/actions-reusable.md
+++ b/docs/actions-reusable.md
@@ -72,10 +72,11 @@ Where *action/cli.py* is:
 
 ```python
 def main():
-  print("A reusable action")
+    print("A reusable action")
+
 
 if __name__ == "__main__":
-  main()
+    main()
 ```
 
 When developing a reusable action, just as when developing a scripted action, the action's dependencies are made available by the runtime; they are not made available by the action.
diff --git a/docs/case-control-studies.md b/docs/case-control-studies.md
index d6bb6c7fc..b7fe62b78 100644
--- a/docs/case-control-studies.md
+++ b/docs/case-control-studies.md
@@ -175,16 +175,17 @@ from ehrql import table_from_file
 CONTROLS = "output/matched_matches.arrow"
 
 matched_patients = table_from_file(
-  CONTROLS,
-  columns={
-    "age": int,
-    "sex": str,
-    "index_date": datetime.date
-  }
+    CONTROLS,
+    columns={
+        "age": int,
+        "sex": str,
+        "index_date": datetime.date,
+    },
 )
 ```
 
 This allows us to only include the matched controls in our dataset.
+<!-- fmt:off -->
 ```python
 from ehrql import create_dataset
 
@@ -193,6 +194,7 @@ dataset.define_population(
     matched_patients.exists_for_patient()
 )
 ```
+<!-- fmt:on -->
 Don't forget to add additional population constraints to the dataset if you require them!
 
 The `index_date` column in `matched_patients` is the date generated by the matching process,
@@ -225,12 +227,12 @@ CONTROLS = "output/matched_matches.arrow"
 codelist = codelist_from_csv("codelists/codelist.csv")
 
 matched_patients = table_from_file(
-  CONTROLS,
-  columns={
-    "age": int,
-    "sex": str,
-    "index_date": datetime.date
-  }
+    CONTROLS,
+    columns={
+        "age": int,
+        "sex": str,
+        "index_date": datetime.date,
+    },
 )
 
 dataset = create_dataset()
diff --git a/docs/federation.md b/docs/federation.md
index ed35d6593..3d01d513d 100644
--- a/docs/federation.md
+++ b/docs/federation.md
@@ -40,16 +40,17 @@ For example, this section of the study definition creates a variable that indica
 
 
 ```python
-
-  sev_obesity = patients.with_these_clinical_events(
-    sev_obesity_codes,
-    returning = "date",
-    ignore_missing_values = True,
-    find_last_match_in_period = True,
-    on_or_after = "bmi_stage_date",
-    on_or_before = "index_date",
-    date_format = "YYYY-MM-DD",
-  ),
+sev_obesity = (
+    patients.with_these_clinical_events(
+        sev_obesity_codes,
+        returning="date",
+        ignore_missing_values=True,
+        find_last_match_in_period=True,
+        on_or_after="bmi_stage_date",
+        on_or_before="index_date",
+        date_format="YYYY-MM-DD",
+    ),
+)
 ```
 
 
diff --git a/hooks/ehrql_branch.py b/hooks/ehrql_branch.py
index 3f2b6dcf9..9e3af0f18 100644
--- a/hooks/ehrql_branch.py
+++ b/hooks/ehrql_branch.py
@@ -19,7 +19,7 @@ def on_config(config):
     ehrql_nav_index, ehrql_nav_section = next(
         (i, section)
         for i, section in enumerate(config["nav"])
-        if list(section.keys())[0] == "ehrQL"
+        if next(iter(section.keys())) == "ehrQL"
     )
     new_import_string = ehrql_nav_section["ehrQL"].replace(
         "branch=main", f"branch={ehrql_branch}"
diff --git a/pyproject.toml b/pyproject.toml
index dffbe0385..5702c6285 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -15,6 +15,8 @@ exclude = [
   "docker",
   "htmlcov",
   "venv",
+  # These are frozen and we don't intend to update them
+  "docs/legacy",
 ]
 
 [tool.ruff.lint]
diff --git a/scripts/__init__.py b/scripts/__init__.py
new file mode 100644
index 000000000..e69de29bb
diff --git a/scripts/wordcount.py b/scripts/wordcount.py
index 1eea5af67..d1e2deae3 100644
--- a/scripts/wordcount.py
+++ b/scripts/wordcount.py
@@ -1,5 +1,5 @@
-from html.parser import HTMLParser
 import sys
+from html.parser import HTMLParser
 
 
 class ArticleParser(HTMLParser):
@@ -24,7 +24,7 @@ def handle_data(self, data):
 if __name__ == "__main__":
     total = 0
     for filename in sys.argv[1:]:
-        with open(filename, "r", encoding="utf-8") as f:
+        with open(filename, encoding="utf-8") as f:
             parser = ArticleParser()
             parser.feed(f.read())
             text = " ".join(parser.text)