feat(urlhaus): new URLhaus malicious-URL intel integration

URLhaus (abuse.ch) API v1, 5 commands: URL/host/payload lookup, recent URLs.
Auth-Key auth, stdlib-only.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

integrations/urlhaus/manifest.yaml

@@ -0,0 +1,63 @@
+id: urlhaus
+name: URLhaus
+version: 1.0.0
+description: "URLhaus by abuse.ch (API v1) — malicious-URL threat intelligence: look up a URL, host, or payload (hash), and pull recently added malicious URLs. Auth-Key authentication; stdlib-only, no extra Python dependencies."
+changelog: "1.0.0 — Initial release: URL/host/payload lookup, recent URLs."
+category: enrichment
+
+# Per-instance configuration. abuse.ch requires an Auth-Key header on API requests.
+config_schema:
+  properties:
+    auth_key:
+      type: string
+      description: "abuse.ch Auth-Key"
+      x-soar-sensitive: true
+  required:
+    - auth_key
+
+commands:
+  - id: url_info
+    name: urlhaus-url-info
+    description: "Look up a URL in the URLhaus database."
+    risk: read
+    inputs_schema:
+      properties:
+        url: { type: string, description: "URL to look up" }
+      required: [url]
+    outputs_schema: { properties: {} }
+  - id: host_info
+    name: urlhaus-host-info
+    description: "Look up a host (domain or IP) in the URLhaus database."
+    risk: read
+    inputs_schema:
+      properties:
+        host: { type: string, description: "Domain or IP" }
+      required: [host]
+    outputs_schema: { properties: {} }
+  - id: payload_info
+    name: urlhaus-payload-info
+    description: "Look up a malware payload by hash (MD5 or SHA-256)."
+    risk: read
+    inputs_schema:
+      properties:
+        hash: { type: string, description: "MD5 or SHA-256 hash" }
+      required: [hash]
+    outputs_schema: { properties: {} }
+  - id: recent_urls
+    name: urlhaus-recent-urls
+    description: "Get recently added malicious URLs."
+    risk: read
+    inputs_schema:
+      properties:
+        limit: { type: string, description: "Result set: 'limit/N' amount — use 'recent' feed (returns latest 1000)" }
+      required: []
+    outputs_schema: { properties: {} }
+
+  - id: test_connection
+    name: urlhaus-test-connection
+    description: "Verify the Auth-Key (used by the Test button)."
+    risk: read
+    inputs_schema:
+      properties: {}
+      required: []
+    outputs_schema: { properties: {} }

integrations/urlhaus/scripts/host_info.py

@@ -0,0 +1,53 @@
+import json, os, sys, urllib.parse, urllib.request, urllib.error
+
+BASE = "https://urlhaus-api.abuse.ch/v1"
+
+
+def _cfg():
+    return json.loads(os.environ.get("INTEGRATION_SECRETS", "{}"))
+
+
+def _inputs():
+    return json.loads(os.environ.get("INTEGRATION_INPUTS", "{}"))
+
+
+def post_form(path, cfg, fields):
+    data = urllib.parse.urlencode({k: v for k, v in fields.items() if v not in (None, "")}).encode("utf-8")
+    headers = {
+        "Auth-Key": str(cfg.get("auth_key", "")),
+        "Content-Type": "application/x-www-form-urlencoded",
+        "Accept": "application/json",
+    }
+    req = urllib.request.Request(BASE + path, data=data, headers=headers, method="POST")
+    with urllib.request.urlopen(req, timeout=60) as r:
+        raw = r.read()
+        return json.loads(raw) if raw else {}
+
+
+def get(path, cfg):
+    headers = {"Auth-Key": str(cfg.get("auth_key", "")), "Accept": "application/json"}
+    req = urllib.request.Request(BASE + path, headers=headers, method="GET")
+    with urllib.request.urlopen(req, timeout=60) as r:
+        raw = r.read()
+        return json.loads(raw) if raw else {}
+
+
+def _run(fn):
+    try:
+        print(json.dumps(fn(_cfg(), _inputs())))
+    except urllib.error.HTTPError as e:
+        print(json.dumps({"error": "HTTP " + str(e.code), "detail": e.read().decode("utf-8", "replace")}))
+        sys.exit(1)
+    except Exception as e:
+        print(json.dumps({"error": str(e)}))
+        sys.exit(1)
+
+
+def main(cfg, inputs):
+    host = str(inputs.get("host", "")).strip()
+    if not host:
+        raise Exception("host is required")
+    return post_form("/host/", cfg, {"host": host})
+
+
+_run(main)

integrations/urlhaus/scripts/payload_info.py

@@ -0,0 +1,54 @@
+import json, os, sys, urllib.parse, urllib.request, urllib.error
+
+BASE = "https://urlhaus-api.abuse.ch/v1"
+
+
+def _cfg():
+    return json.loads(os.environ.get("INTEGRATION_SECRETS", "{}"))
+
+
+def _inputs():
+    return json.loads(os.environ.get("INTEGRATION_INPUTS", "{}"))
+
+
+def post_form(path, cfg, fields):
+    data = urllib.parse.urlencode({k: v for k, v in fields.items() if v not in (None, "")}).encode("utf-8")
+    headers = {
+        "Auth-Key": str(cfg.get("auth_key", "")),
+        "Content-Type": "application/x-www-form-urlencoded",
+        "Accept": "application/json",
+    }
+    req = urllib.request.Request(BASE + path, data=data, headers=headers, method="POST")
+    with urllib.request.urlopen(req, timeout=60) as r:
+        raw = r.read()
+        return json.loads(raw) if raw else {}
+
+
+def get(path, cfg):
+    headers = {"Auth-Key": str(cfg.get("auth_key", "")), "Accept": "application/json"}
+    req = urllib.request.Request(BASE + path, headers=headers, method="GET")
+    with urllib.request.urlopen(req, timeout=60) as r:
+        raw = r.read()
+        return json.loads(raw) if raw else {}
+
+
+def _run(fn):
+    try:
+        print(json.dumps(fn(_cfg(), _inputs())))
+    except urllib.error.HTTPError as e:
+        print(json.dumps({"error": "HTTP " + str(e.code), "detail": e.read().decode("utf-8", "replace")}))
+        sys.exit(1)
+    except Exception as e:
+        print(json.dumps({"error": str(e)}))
+        sys.exit(1)
+
+
+def main(cfg, inputs):
+    hash_ = str(inputs.get("hash", "")).strip()
+    if not hash_:
+        raise Exception("hash is required")
+    field = "md5_hash" if len(hash_) == 32 else "sha256_hash"
+    return post_form("/payload/", cfg, {field: hash_})
+
+
+_run(main)

integrations/urlhaus/scripts/recent_urls.py

@@ -0,0 +1,50 @@
+import json, os, sys, urllib.parse, urllib.request, urllib.error
+
+BASE = "https://urlhaus-api.abuse.ch/v1"
+
+
+def _cfg():
+    return json.loads(os.environ.get("INTEGRATION_SECRETS", "{}"))
+
+
+def _inputs():
+    return json.loads(os.environ.get("INTEGRATION_INPUTS", "{}"))
+
+
+def post_form(path, cfg, fields):
+    data = urllib.parse.urlencode({k: v for k, v in fields.items() if v not in (None, "")}).encode("utf-8")
+    headers = {
+        "Auth-Key": str(cfg.get("auth_key", "")),
+        "Content-Type": "application/x-www-form-urlencoded",
+        "Accept": "application/json",
+    }
+    req = urllib.request.Request(BASE + path, data=data, headers=headers, method="POST")
+    with urllib.request.urlopen(req, timeout=60) as r:
+        raw = r.read()
+        return json.loads(raw) if raw else {}
+
+
+def get(path, cfg):
+    headers = {"Auth-Key": str(cfg.get("auth_key", "")), "Accept": "application/json"}
+    req = urllib.request.Request(BASE + path, headers=headers, method="GET")
+    with urllib.request.urlopen(req, timeout=60) as r:
+        raw = r.read()
+        return json.loads(raw) if raw else {}
+
+
+def _run(fn):
+    try:
+        print(json.dumps(fn(_cfg(), _inputs())))
+    except urllib.error.HTTPError as e:
+        print(json.dumps({"error": "HTTP " + str(e.code), "detail": e.read().decode("utf-8", "replace")}))
+        sys.exit(1)
+    except Exception as e:
+        print(json.dumps({"error": str(e)}))
+        sys.exit(1)
+
+
+def main(cfg, inputs):
+    return get("/urls/recent/", cfg)
+
+
+_run(main)

integrations/urlhaus/scripts/test_connection.py

@@ -0,0 +1,51 @@
+import json, os, sys, urllib.parse, urllib.request, urllib.error
+
+BASE = "https://urlhaus-api.abuse.ch/v1"
+
+
+def _cfg():
+    return json.loads(os.environ.get("INTEGRATION_SECRETS", "{}"))
+
+
+def _inputs():
+    return json.loads(os.environ.get("INTEGRATION_INPUTS", "{}"))
+
+
+def post_form(path, cfg, fields):
+    data = urllib.parse.urlencode({k: v for k, v in fields.items() if v not in (None, "")}).encode("utf-8")
+    headers = {
+        "Auth-Key": str(cfg.get("auth_key", "")),
+        "Content-Type": "application/x-www-form-urlencoded",
+        "Accept": "application/json",
+    }
+    req = urllib.request.Request(BASE + path, data=data, headers=headers, method="POST")
+    with urllib.request.urlopen(req, timeout=60) as r:
+        raw = r.read()
+        return json.loads(raw) if raw else {}
+
+
+def get(path, cfg):
+    headers = {"Auth-Key": str(cfg.get("auth_key", "")), "Accept": "application/json"}
+    req = urllib.request.Request(BASE + path, headers=headers, method="GET")
+    with urllib.request.urlopen(req, timeout=60) as r:
+        raw = r.read()
+        return json.loads(raw) if raw else {}
+
+
+def _run(fn):
+    try:
+        print(json.dumps(fn(_cfg(), _inputs())))
+    except urllib.error.HTTPError as e:
+        print(json.dumps({"error": "HTTP " + str(e.code), "detail": e.read().decode("utf-8", "replace")}))
+        sys.exit(1)
+    except Exception as e:
+        print(json.dumps({"error": str(e)}))
+        sys.exit(1)
+
+
+def main(cfg, inputs):
+    get("/urls/recent/", cfg)
+    return {"ok": True}
+
+
+_run(main)

integrations/urlhaus/scripts/url_info.py

@@ -0,0 +1,53 @@
+import json, os, sys, urllib.parse, urllib.request, urllib.error
+
+BASE = "https://urlhaus-api.abuse.ch/v1"
+
+
+def _cfg():
+    return json.loads(os.environ.get("INTEGRATION_SECRETS", "{}"))
+
+
+def _inputs():
+    return json.loads(os.environ.get("INTEGRATION_INPUTS", "{}"))
+
+
+def post_form(path, cfg, fields):
+    data = urllib.parse.urlencode({k: v for k, v in fields.items() if v not in (None, "")}).encode("utf-8")
+    headers = {
+        "Auth-Key": str(cfg.get("auth_key", "")),
+        "Content-Type": "application/x-www-form-urlencoded",
+        "Accept": "application/json",
+    }
+    req = urllib.request.Request(BASE + path, data=data, headers=headers, method="POST")
+    with urllib.request.urlopen(req, timeout=60) as r:
+        raw = r.read()
+        return json.loads(raw) if raw else {}
+
+
+def get(path, cfg):
+    headers = {"Auth-Key": str(cfg.get("auth_key", "")), "Accept": "application/json"}
+    req = urllib.request.Request(BASE + path, headers=headers, method="GET")
+    with urllib.request.urlopen(req, timeout=60) as r:
+        raw = r.read()
+        return json.loads(raw) if raw else {}
+
+
+def _run(fn):
+    try:
+        print(json.dumps(fn(_cfg(), _inputs())))
+    except urllib.error.HTTPError as e:
+        print(json.dumps({"error": "HTTP " + str(e.code), "detail": e.read().decode("utf-8", "replace")}))
+        sys.exit(1)
+    except Exception as e:
+        print(json.dumps({"error": str(e)}))
+        sys.exit(1)
+
+
+def main(cfg, inputs):
+    url = str(inputs.get("url", "")).strip()
+    if not url:
+        raise Exception("url is required")
+    return post_form("/url/", cfg, {"url": url})
+
+
+_run(main)