From 65cbc6922c01472e92afb5e7e72ba7e768614160 Mon Sep 17 00:00:00 2001
From: Xe Iaso <me@xeiaso.net>
Date: Thu, 24 Apr 2025 19:15:20 -0400
Subject: [PATCH] data: add some examples of expression-based rules

Signed-off-by: Xe Iaso <me@xeiaso.net>
---
 data/apps/git-client.yaml              | 7 +++++++
 data/clients/go-get.yaml               | 7 +++++++
 data/common/allow-api-like.yaml        | 6 ++++++
 data/common/keep-internet-working.yaml | 3 +++
 data/common/rfc-violations.yaml        | 3 +++
 data/embed.go                          | 2 +-
 6 files changed, 27 insertions(+), 1 deletion(-)
 create mode 100644 data/apps/git-client.yaml
 create mode 100644 data/clients/go-get.yaml
 create mode 100644 data/common/allow-api-like.yaml
 create mode 100644 data/common/rfc-violations.yaml

diff --git a/data/apps/git-client.yaml b/data/apps/git-client.yaml
new file mode 100644
index 00000000..b2964937
--- /dev/null
+++ b/data/apps/git-client.yaml
@@ -0,0 +1,7 @@
+- name: allow-git-clients
+  action: ALLOW
+  expression:
+    and:
+    - userAgent.startsWith("git/") || userAgent.contains("libgit") || userAgent.startsWith("go-git") || userAgent.startsWith("JGit/") || userAgent.startsWith("JGit-")
+    - >
+      "Git-Protocol" in headers && headers["Git-Protocol"] == "version=2"
\ No newline at end of file
diff --git a/data/clients/go-get.yaml b/data/clients/go-get.yaml
new file mode 100644
index 00000000..ca061e28
--- /dev/null
+++ b/data/clients/go-get.yaml
@@ -0,0 +1,7 @@
+- name: go-get
+  action: ALLOW
+  expression:
+    and:
+    - userAgent.startsWith("Go-http-client/")
+    - '"go-get" in query'
+    - query["go-get"] == "1"
\ No newline at end of file
diff --git a/data/common/allow-api-like.yaml b/data/common/allow-api-like.yaml
new file mode 100644
index 00000000..50738123
--- /dev/null
+++ b/data/common/allow-api-like.yaml
@@ -0,0 +1,6 @@
+- name: allow-api-routes
+  action: ALLOW
+  expression:
+    and:
+    - '!(method == "HEAD" || method == "GET")'
+    - path.startsWith("/api/")
\ No newline at end of file
diff --git a/data/common/keep-internet-working.yaml b/data/common/keep-internet-working.yaml
index 8270ef4e..d72f5abf 100644
--- a/data/common/keep-internet-working.yaml
+++ b/data/common/keep-internet-working.yaml
@@ -7,4 +7,7 @@
   action: ALLOW
 - name: robots-txt
   path_regex: ^/robots.txt$
+  action: ALLOW
+- name: sitemap
+  path_regex: ^/sitemap.xml$
   action: ALLOW
\ No newline at end of file
diff --git a/data/common/rfc-violations.yaml b/data/common/rfc-violations.yaml
new file mode 100644
index 00000000..77ffe9eb
--- /dev/null
+++ b/data/common/rfc-violations.yaml
@@ -0,0 +1,3 @@
+- name: no-user-agent-string
+  expression: userAgent == ""
+  action: DENY
\ No newline at end of file
diff --git a/data/embed.go b/data/embed.go
index ebb2152f..06cf30d3 100644
--- a/data/embed.go
+++ b/data/embed.go
@@ -3,6 +3,6 @@ package data
 import "embed"
 
 var (
-	//go:embed botPolicies.yaml botPolicies.json apps bots common crawlers
+	//go:embed botPolicies.yaml botPolicies.json apps bots clients common crawlers
 	BotPolicies embed.FS
 )