diff --git a/.github/workflows/shadow.yml b/.github/workflows/shadow.yml
new file mode 100644
index 00000000000..06eaff37cf7
--- /dev/null
+++ b/.github/workflows/shadow.yml
@@ -0,0 +1,47 @@
+# Shadow PR review bot — installed via reusable workflow at sudsali/shadow.
+# See https://github.com/sudsali/shadow for docs and security model.
+name: Shadow
+
+# Embed the PR/issue number in the run name. The reusable workflow's run-name
+# is ignored when invoked via workflow_call — only the caller's template
+# applies. This is what lets the rate-limit gate (BOT_MAX_RUNS_PER_HOUR) match
+# prior runs for issue/issue_comment events, where the workflow_run record
+# has no `pull_requests` array to match on.
+run-name: "Shadow #${{ github.event.pull_request.number || github.event.issue.number || inputs.pr_number }}"
+
+on:
+  pull_request_target:
+    types: [opened, reopened, synchronize]
+  issues:
+    types: [opened, reopened]
+  issue_comment:
+    types: [created]
+  workflow_dispatch:
+    inputs:
+      pr_number:
+        required: true
+        description: "PR number to review"
+      dry_run:
+        type: boolean
+        default: false
+      shadow_ref:
+        default: v0
+        description: "Pin to a Shadow release (v0 or commit SHA)"
+
+permissions:
+  contents: read
+  id-token: write
+  pull-requests: write
+  issues: write
+
+jobs:
+  shadow:
+    uses: sudsali/shadow/.github/workflows/shadow-review.yml@v0
+    with:
+      pr_number: ${{ inputs.pr_number }}
+      dry_run: ${{ inputs.dry_run && 'true' || 'false' }}
+      shadow_ref: ${{ inputs.shadow_ref || 'v0' }}
+    secrets:
+      AWS_ROLE_ARN: ${{ secrets.AWS_ROLE_ARN }}
+      GUARDRAIL_ID: ${{ secrets.GUARDRAIL_ID }}
+      GUARDRAIL_VERSION: ${{ secrets.GUARDRAIL_VERSION }}
diff --git a/.shadow.yml b/.shadow.yml
new file mode 100644
index 00000000000..1975694c68a
--- /dev/null
+++ b/.shadow.yml
@@ -0,0 +1,15 @@
+# Shadow PR review bot config — see https://github.com/sudsali/shadow
+codebase:
+  src_dir: aws_lambda_powertools     # primary source root
+  file_ext: .py                      # narrow tool searches to Python
+  test_dir: tests                    # used by find_tests_for
+  language: python                   # legacy issue-respond path only
+
+bot:
+  name: shadow
+  escalate_label: needs-human
+
+models:
+  investigator: us.anthropic.claude-opus-4-7
+  critic: us.anthropic.claude-opus-4-7
+  reporter: us.anthropic.claude-haiku-4-5-20251001-v1:0