Split up AI filtering files (#592)

* Split up AI filtering files

Create aggressive/moderate/permissive policies to allow administrators to choose their AI/LLM stance.

Aggressive policy matches existing default in Anubis.

Removes `Google-Extended` flag from `ai-robots-txt.yaml` as it doesn't exist in requests.

Rename `ai-robots-txt.yaml` to `ai-catchall.yaml` as the file is no longer a copy of the source repo/file.

* chore: spelling

* chore: fix embeds

* chore: fix data includes

* chore: fix file name typo

* chore: Ignore READMEs in configs

* chore(lib/policy/config): go tool goimports -w

Signed-off-by: Xe Iaso <me@xeiaso.net>

---------

Signed-off-by: Xe Iaso <me@xeiaso.net>
Co-authored-by: Xe Iaso <me@xeiaso.net>

data/botPolicies.json

@@ -4,7 +4,7 @@
       "import": "(data)/bots/_deny-pathological.yaml"
     },
     {
-      "import": "(data)/bots/ai-robots-txt.yaml"
+      "import": "(data)/meta/ai-block-aggressive.yaml"
     },
     {
       "import": "(data)/crawlers/_allow-good.yaml"

data/botPolicies.yaml

@@ -17,8 +17,12 @@ bots:
     import: (data)/bots/_deny-pathological.yaml
   - import: (data)/bots/aggressive-brazilian-scrapers.yaml
 
-  # Enforce https://github.com/ai-robots-txt/ai.robots.txt
-  - import: (data)/bots/ai-robots-txt.yaml
+  # Aggressively block AI/LLM related bots/agents by default
+  - import: (data)/meta/ai-block-aggressive.yaml
+
+  # Consider replacing the aggressive AI policy with more selective policies:
+  # - import: (data)/meta/ai-block-moderate.yaml
+  # - import: (data)/meta/ai-block-permissive.yaml
 
   # Search engine crawlers to allow, defaults to:
   #   - Google (so they don't try to bypass Anubis)

data/bots/ai-catchall.yaml

@@ -0,0 +1,11 @@
+# Extensive list of AI-affiliated agents based on https://github.com/ai-robots-txt/ai.robots.txt
+# Add new/undocumented agents here. Where documentation exists, consider moving to dedicated policy files.
+# Notes on various agents:
+#  - Amazonbot: Well documented, but they refuse to state which agent collects training data.
+#  - anthropic-ai/Claude-Web: Undocumented by Anthropic. Possibly deprecated or hallucinations?
+#  - Perplexity*: Well documented, but they refuse to state which agent collects training data.
+# Warning: May contain user agents that _must_ be blocked in robots.txt, or the opt-out will have no effect.
+- name: "ai-catchall"
+  user_agent_regex: >-
+    AI2Bot|Ai2Bot-Dolma|aiHitBot|Amazonbot|anthropic-ai|Brightbot 1.0|Bytespider|CCBot|Claude-Web|cohere-ai|cohere-training-data-crawler|Cotoyogi|Crawlspace|Diffbot|DuckAssistBot|FacebookBot|Factset_spyderbot|FirecrawlAgent|FriendlyCrawler|Google-CloudVertexBot|GoogleOther|GoogleOther-Image|GoogleOther-Video|iaskspider/2.0|ICC-Crawler|ImagesiftBot|img2dataset|imgproxy|ISSCyberRiskCrawler|Kangaroo Bot|meta-externalagent|Meta-ExternalAgent|meta-externalfetcher|Meta-ExternalFetcher|NovaAct|omgili|omgilibot|Operator|PanguBot|Perplexity-User|PerplexityBot|PetalBot|QualifiedBot|Scrapy|SemrushBot-OCOB|SemrushBot-SWA|Sidetrade indexer bot|TikTokSpider|Timpibot|VelenPublicWebCrawler|Webzio-Extended|wpbot|YouBot
+  action: DENY

data/bots/ai-robots-txt.yaml

@@ -1,6 +0,0 @@
-# Warning: Contains user agents that _must_ be blocked in robots.txt, or the opt-out will have no effect.
-# Note: Blocks human-directed/non-training user agents
-- name: "ai-robots-txt"
-  user_agent_regex: >-
-    AI2Bot|Ai2Bot-Dolma|aiHitBot|Amazonbot|anthropic-ai|Brightbot 1.0|Bytespider|CCBot|ChatGPT-User|Claude-SearchBot|Claude-User|Claude-Web|ClaudeBot|cohere-ai|cohere-training-data-crawler|Cotoyogi|Crawlspace|Diffbot|DuckAssistBot|FacebookBot|Factset_spyderbot|FirecrawlAgent|FriendlyCrawler|Google-CloudVertexBot|Google-Extended|GoogleOther|GoogleOther-Image|GoogleOther-Video|GPTBot|iaskspider/2.0|ICC-Crawler|ImagesiftBot|img2dataset|imgproxy|ISSCyberRiskCrawler|Kangaroo Bot|meta-externalagent|Meta-ExternalAgent|meta-externalfetcher|Meta-ExternalFetcher|MistralAI-User/1.0|NovaAct|OAI-SearchBot|omgili|omgilibot|Operator|PanguBot|Perplexity-User|PerplexityBot|PetalBot|QualifiedBot|Scrapy|SemrushBot-OCOB|SemrushBot-SWA|Sidetrade indexer bot|TikTokSpider|Timpibot|VelenPublicWebCrawler|Webzio-Extended|wpbot|YouBot
-  action: DENY

data/clients/ai.yaml

@@ -0,0 +1,8 @@
+# User agents that act on behalf of humans in AI tools, e.g. searching the web.
+# Each entry should have a positive/ALLOW entry created as well, with further documentation.
+# Exceptions:
+#  - Claude-User: No published IP allowlist
+- name: "ai-clients"
+  user_agent_regex: >-
+    ChatGPT-User|Claude-User|MistralAI-User
+  action: DENY

data/crawlers/ai-search.yaml

@@ -0,0 +1,8 @@
+# User agents that index exclusively for search in for AI systems.
+# Each entry should have a positive/ALLOW entry created as well, with further documentation.
+# Exceptions:
+#  - Claude-SearchBot: No published IP allowlist
+- name: "ai-crawlers-search"
+  user_agent_regex: >-
+    OAI-SearchBot|Claude-SearchBot
+  action: DENY

data/crawlers/ai-training.yaml

@@ -0,0 +1,8 @@
+# User agents that crawl for training AI/LLM systems
+# Each entry should have a positive/ALLOW entry created as well, with further documentation.
+# Exceptions:
+#  - ClaudeBot: No published IP allowlist
+- name: "ai-crawlers-training"
+  user_agent_regex: >-
+    GPTBot|ClaudeBot
+  action: DENY

data/embed.go

@@ -3,6 +3,6 @@ package data
 import "embed"
 
 var (
-	//go:embed botPolicies.yaml botPolicies.json all:apps all:bots all:clients all:common all:crawlers
+	//go:embed botPolicies.yaml botPolicies.json all:apps all:bots all:clients all:common all:crawlers all:meta
 	BotPolicies embed.FS
 )

data/meta/README.md

@@ -0,0 +1,5 @@
+# meta policies
+
+Contains policies that exclusively reference policies in _multiple_ other data folders.
+
+Akin to "stances" that the administrator can take, with reference to various topics, such as AI/LLM systems.

data/meta/ai-block-aggressive.yaml

@@ -0,0 +1,6 @@
+# Blocks all AI/LLM associated user agents, regardless of purpose or human agency
+# Warning: To completely block some AI/LLM training, such as with Google, you _must_ place flags in robots.txt.
+- import: (data)/bots/ai-catchall.yaml
+- import: (data)/clients/ai.yaml
+- import: (data)/crawlers/ai-search.yaml
+- import: (data)/crawlers/ai-training.yaml

data/meta/ai-block-moderate.yaml

@@ -0,0 +1,7 @@
+# Blocks all AI/LLM bots used for training or unknown/undocumented purposes.
+# Permits user agents with explicitly documented non-training use, and published IP allowlists.
+- import: (data)/bots/ai-catchall.yaml
+- import: (data)/crawlers/ai-training.yaml
+- import: (data)/crawlers/openai-searchbot.yaml
+- import: (data)/clients/openai-chatgpt-user.yaml
+- import: (data)/clients/mistral-mistralai-user.yaml

data/meta/ai-block-permissive.yaml

@@ -0,0 +1,6 @@
+# Permits all well documented AI/LLM user agents with published IP allowlists.
+- import: (data)/bots/ai-catchall.yaml
+- import: (data)/crawlers/openai-searchbot.yaml
+- import: (data)/crawlers/openai-gptbot.yaml
+- import: (data)/clients/openai-chatgpt-user.yaml
+- import: (data)/clients/mistral-mistralai-user.yaml