govdata-mcp 0.1.0__tar.gz

govdata_mcp-0.1.0/.github/workflows/publish.yml

@@ -0,0 +1,20 @@
+name: Publish to PyPI
+
+on:
+  push:
+    tags: ["v*"]
+
+jobs:
+  publish:
+    runs-on: ubuntu-latest
+    permissions:
+      id-token: write # required for PyPI trusted publishing (OIDC)
+      contents: read
+    steps:
+      - uses: actions/checkout@v4
+      - uses: actions/setup-python@v5
+        with:
+          python-version: "3.12"
+      - run: pip install build
+      - run: python -m build
+      - uses: pypa/gh-action-pypi-publish@release/v1

govdata_mcp-0.1.0/.github/workflows/registry.yml

@@ -0,0 +1,61 @@
+name: Publish to MCP Registry
+
+# Official MCP Registry (registry.modelcontextprotocol.io) via GitHub OIDC.
+# No secret, no interactive login: the runner's OIDC token proves ownership
+# of the io.github.mohammad-abbasi-sh/* namespace.
+# Refs: https://github.com/modelcontextprotocol/registry
+
+on:
+  push:
+    tags: ["v*"]
+  workflow_dispatch:
+
+jobs:
+  publish:
+    runs-on: ubuntu-latest
+    permissions:
+      id-token: write # required for OIDC -> MCP Registry
+      contents: read
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Sync server.json version to tag
+        # Only sync when running on a version tag (v*). On workflow_dispatch
+        # (ref = refs/heads/main) keep the version committed in server.json.
+        run: |
+          VER=""
+          if [[ "$GITHUB_REF" == refs/tags/* ]]; then
+            TAG="${GITHUB_REF#refs/tags/}"
+            VER="${TAG#v}"
+          fi
+          if [ -n "$VER" ]; then
+            python3 - "$VER" << 'EOF'
+          import json, sys
+          ver = sys.argv[1]
+          d = json.load(open('server.json'))
+          d['version'] = ver
+          for pkg in d.get('packages', []):
+              pkg['version'] = ver
+          json.dump(d, open('server.json', 'w'), indent=2, ensure_ascii=False)
+          print('server.json version ->', ver)
+          EOF
+          else
+            echo "No version tag ($GITHUB_REF); keeping server.json version as-is"
+          fi
+
+      - name: Install mcp-publisher
+        run: |
+          curl -sL "https://github.com/modelcontextprotocol/registry/releases/download/v1.8.7/mcp-publisher_linux_amd64.tar.gz" -o /tmp/mcp-publisher.tar.gz
+          tar -xzf /tmp/mcp-publisher.tar.gz -C /tmp
+          sudo mv /tmp/mcp-publisher /usr/local/bin/mcp-publisher
+          mcp-publisher --version
+
+      - name: Validate server.json
+        run: mcp-publisher validate server.json
+
+      - name: Publish to MCP Registry (OIDC)
+        run: |
+          mcp-publisher login github-oidc
+          mcp-publisher publish server.json
+        # A registry outage must never break the release; re-run from Actions if needed.
+        continue-on-error: true

govdata_mcp-0.1.0/.github/workflows/smoke.yml

@@ -0,0 +1,21 @@
+name: Daily live smoke
+
+# One fixed probe per day: catches fingerprint rot / bot-wall changes
+# before users hit them. Fails visibly on 403/timeout.
+on:
+  schedule:
+    - cron: "0 6 * * *"
+  workflow_dispatch:
+
+jobs:
+  smoke:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+      - uses: actions/setup-python@v5
+        with:
+          python-version: "3.12"
+      - name: Install
+        run: pip install -e .
+      - name: Smoke probe
+        run: python tests/smoke_live.py

govdata_mcp-0.1.0/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 Mohammad Abbasi
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

govdata_mcp-0.1.0/PKG-INFO

@@ -0,0 +1,70 @@
+Metadata-Version: 2.5
+Name: govdata-mcp
+Version: 0.1.0
+Summary: Unofficial read-only MCP server for govdata.de (German open data, CKAN) — search datasets, publishers, resources
+Author: Mohammad Abbasi
+License: MIT
+License-File: LICENSE
+Requires-Python: >=3.10
+Requires-Dist: httpx>=0.27
+Requires-Dist: mcp<2
+Description-Content-Type: text/markdown
+
+# govdata-mcp 🗃️
+
+mcp-name: io.github.mohammad-abbasi-sh/govdata-mcp
+
+Unofficial **read-only** MCP server for **[govdata.de](https://www.govdata.de)** — Germany's national open-data portal govdata.de via its official CKAN API: search 163,000+ datasets, dataset details, publishers, recent datasets and resource details. No key, no login. No key, no login.
+
+> Built by Mohammad Abbasi as part of the "Germany MCP factory". Not affiliated with govdata.de.
+
+## For AI agents
+
+Add to your MCP client config (stdio):
+
+```json
+{
+  "mcpServers": {
+    "govdata": {
+      "command": "uvx",
+      "args": ["govdata-mcp==0.1.0"]
+    }
+  }
+}
+```
+
+Or run locally: `python -m govdata_mcp.server` (needs `mcp<2`, `httpx`).
+
+## Tools
+
+| Tool | Description |
+|------|-------------|
+| `govdata_search_datasets` | Search govdata.de datasets — `query`, `limit` 1..50, `sort` (relevance/recent) |
+| `govdata_get_dataset` | Full details of one dataset — `dataset_id` (id or name, e.g. "wohnatlas-berlin") |
+| `govdata_list_publishers` | Publishing organisations on govdata.de, most datasets first — `limit` 1..50 |
+| `govdata_recent_datasets` | Most recently updated datasets — `limit` 1..50 |
+| `govdata_get_resource` | Details of one dataset resource (file/API link) — `resource_id` |
+| `govdata_health` | Reachability probe (ok / blocked / error) — one polite request |
+
+## Notes
+
+- All data comes from the **govdata.de CKAN API** (free, no key, no login) — 163,247 datasets at last check.
+- Competition note: pipeworx-io/mcp-govdata-de (★0, 2026-11-09) covers the same portal but only ships 2 tools (`list_organizations`, `list_groups`) behind a commercial gateway and has no dataset search; this server adds search, dataset/resource details, recency and tests.
+
+Tests: 19/19 green (13 offline + 6 live, 2026-11-10).
+
+## Development
+
+```bash
+python3 -m venv .venv && .venv/bin/pip install "mcp<2" httpx pytest
+.venv/bin/pip install -e .
+.venv/bin/python -m pytest tests/ -q
+```
+
+## License
+
+MIT
+
+---
+
+Built by [Mohammad Abbasi](https://github.com/mohammad-abbasi-sh) — [Almanibama](https://almanibama.ir) · [mabbasi.ir](https://mabbasi.ir)

govdata_mcp-0.1.0/README-de.md

@@ -0,0 +1,51 @@
+# govdata-mcp 🗃️
+
+mcp-name: io.github.mohammad-abbasi-sh/govdata-mcp
+
+Inoffizieller **nur lesender** MCP-Server für **[govdata.de](https://www.govdata.de)** — Deutschlands nationales Open-Data-Portal govdata.de über die offizielle CKAN-API: über 163.000 Datensätze durchsuchen, Datensatzdetails, Herausgeber, neueste Datensätze und Ressourcendetails. Kein Schlüssel, kein Login. Kein Schlüssel, keine Anmeldung.
+
+> Gebaut von Mohammad Abbasi als Teil der "Germany MCP factory". Nicht mit govdata.de verbunden.
+
+## Für KI-Agenten
+
+Zur MCP-Client-Konfiguration hinzufügen (stdio):
+
+```json
+{
+  "mcpServers": {
+    "govdata": {
+      "command": "uvx",
+      "args": ["govdata-mcp==0.1.0"]
+    }
+  }
+}
+```
+
+Oder lokal starten: `python -m govdata_mcp.server` (benötigt `mcp<2`, `httpx`).
+
+## Tools
+
+| Tool | Beschreibung |
+|------|-------------|
+| `govdata_search_datasets` | Datensätze auf govdata.de suchen |
+| `govdata_get_dataset` | Vollständige Details eines Datensatzes |
+| `govdata_list_publishers` | Herausgeber, nach Datensatzzahl sortiert |
+| `govdata_recent_datasets` | Zuletzt aktualisierte Datensätze |
+| `govdata_get_resource` | Details einer Datensatz-Ressource |
+| `govdata_health` | Erreichbarkeitsprüfung (ok / blocked / error) — eine höfliche Anfrage |
+
+## Entwicklung
+
+```bash
+python3 -m venv .venv && .venv/bin/pip install "mcp<2" httpx pytest
+.venv/bin/pip install -e .
+.venv/bin/python -m pytest tests/ -q
+```
+
+## Lizenz
+
+MIT
+
+---
+
+Built by [Mohammad Abbasi](https://github.com/mohammad-abbasi-sh) — [Almanibama](https://almanibama.ir) · [mabbasi.ir](https://mabbasi.ir)

govdata_mcp-0.1.0/README-fa.md

@@ -0,0 +1,51 @@
+# govdata-mcp 🗃️
+
+mcp-name: io.github.mohammad-abbasi-sh/govdata-mcp
+
+سرور MCP غیررسمی و **فقط‌خواندنی** برای **[govdata.de](https://www.govdata.de)** — پورتال ملی داده‌های باز آلمان govdata.de از طریق API رسمی CKAN: جست‌وجو در بیش از ۱۶۳٬۰۰۰ مجموعه‌داده، جزئیات مجموعه‌داده، ناشران، تازه‌ترین مجموعه‌داده‌ها و جزئیات منبع. بدون کلید، بدون ورود. بدون کلید، بدون ورود.
+
+> ساختهٔ محمد عباسی، بخشی از «کارخانهٔ MCP آلمان». وابسته به govdata.de نیست.
+
+## برای ایجنت‌های هوش مصنوعی
+
+به پیکربندی MCP کلاینت اضافه کنید (stdio):
+
+```json
+{
+  "mcpServers": {
+    "govdata": {
+      "command": "uvx",
+      "args": ["govdata-mcp==0.1.0"]
+    }
+  }
+}
+```
+
+یا به‌صورت محلی اجرا کنید: `python -m govdata_mcp.server` (نیازمند `mcp<2` و `httpx`).
+
+## ابزارها
+
+| ابزار | توضیح |
+|------|-------|
+| `govdata_search_datasets` | جست‌وجوی مجموعه‌داده‌ها در govdata.de |
+| `govdata_get_dataset` | جزئیات کامل یک مجموعه‌داده |
+| `govdata_list_publishers` | ناشران، به ترتیب بیشترین مجموعه‌داده |
+| `govdata_recent_datasets` | تازه‌ترین مجموعه‌داده‌های به‌روز شده |
+| `govdata_get_resource` | جزئیات یک منبع مجموعه‌داده |
+| `govdata_health` | بررسی دسترس‌پذیری (ok / blocked / error) — یک درخواست مؤدبانه |
+
+## توسعه
+
+```bash
+python3 -m venv .venv && .venv/bin/pip install "mcp<2" httpx pytest
+.venv/bin/pip install -e .
+.venv/bin/python -m pytest tests/ -q
+```
+
+## لایسنس
+
+MIT
+
+---
+
+Built by [Mohammad Abbasi](https://github.com/mohammad-abbasi-sh) — [Almanibama](https://almanibama.ir) · [mabbasi.ir](https://mabbasi.ir)

govdata_mcp-0.1.0/README.md

@@ -0,0 +1,58 @@
+# govdata-mcp 🗃️
+
+mcp-name: io.github.mohammad-abbasi-sh/govdata-mcp
+
+Unofficial **read-only** MCP server for **[govdata.de](https://www.govdata.de)** — Germany's national open-data portal govdata.de via its official CKAN API: search 163,000+ datasets, dataset details, publishers, recent datasets and resource details. No key, no login. No key, no login.
+
+> Built by Mohammad Abbasi as part of the "Germany MCP factory". Not affiliated with govdata.de.
+
+## For AI agents
+
+Add to your MCP client config (stdio):
+
+```json
+{
+  "mcpServers": {
+    "govdata": {
+      "command": "uvx",
+      "args": ["govdata-mcp==0.1.0"]
+    }
+  }
+}
+```
+
+Or run locally: `python -m govdata_mcp.server` (needs `mcp<2`, `httpx`).
+
+## Tools
+
+| Tool | Description |
+|------|-------------|
+| `govdata_search_datasets` | Search govdata.de datasets — `query`, `limit` 1..50, `sort` (relevance/recent) |
+| `govdata_get_dataset` | Full details of one dataset — `dataset_id` (id or name, e.g. "wohnatlas-berlin") |
+| `govdata_list_publishers` | Publishing organisations on govdata.de, most datasets first — `limit` 1..50 |
+| `govdata_recent_datasets` | Most recently updated datasets — `limit` 1..50 |
+| `govdata_get_resource` | Details of one dataset resource (file/API link) — `resource_id` |
+| `govdata_health` | Reachability probe (ok / blocked / error) — one polite request |
+
+## Notes
+
+- All data comes from the **govdata.de CKAN API** (free, no key, no login) — 163,247 datasets at last check.
+- Competition note: pipeworx-io/mcp-govdata-de (★0, 2026-11-09) covers the same portal but only ships 2 tools (`list_organizations`, `list_groups`) behind a commercial gateway and has no dataset search; this server adds search, dataset/resource details, recency and tests.
+
+Tests: 19/19 green (13 offline + 6 live, 2026-11-10).
+
+## Development
+
+```bash
+python3 -m venv .venv && .venv/bin/pip install "mcp<2" httpx pytest
+.venv/bin/pip install -e .
+.venv/bin/python -m pytest tests/ -q
+```
+
+## License
+
+MIT
+
+---
+
+Built by [Mohammad Abbasi](https://github.com/mohammad-abbasi-sh) — [Almanibama](https://almanibama.ir) · [mabbasi.ir](https://mabbasi.ir)

govdata_mcp-0.1.0/pyproject.toml

@@ -0,0 +1,25 @@
+[build-system]
+requires = ["hatchling"]
+build-backend = "hatchling.build"
+
+[project]
+name = "govdata-mcp"
+version = "0.1.0"
+description = "Unofficial read-only MCP server for govdata.de (German open data, CKAN) — search datasets, publishers, resources"
+readme = "README.md"
+license = { text = "MIT" }
+requires-python = ">=3.10"
+dependencies = [
+    "mcp<2",
+    "httpx>=0.27",
+]
+authors = [{ name = "Mohammad Abbasi" }]
+
+[project.scripts]
+govdata-mcp = "govdata_mcp.server:main"
+
+[tool.hatch.build.targets.wheel]
+packages = ["src/govdata_mcp"]
+
+[tool.pytest.ini_options]
+markers = ["live: tests that hit the real https://www.govdata.de site"]

govdata_mcp-0.1.0/server.json

@@ -0,0 +1 @@
+{"$schema": "https://static.modelcontextprotocol.io/schemas/2025-12-11/server.schema.json", "description": "Unofficial read-only MCP server for govdata.de open data (CKAN) — search datasets & publishers.", "name": "io.github.mohammad-abbasi-sh/govdata-mcp", "packages": [{"identifier": "govdata-mcp", "registryBaseUrl": "https://pypi.org", "registryType": "pypi", "transport": {"type": "stdio"}, "version": "0.1.0"}], "repository": {"source": "github", "url": "https://github.com/mohammad-abbasi-sh/govdata-mcp"}, "title": "GovData — Germany's open-data portal: search 163,000+ datasets, publishers & resources", "version": "0.1.0", "websiteUrl": "https://github.com/mohammad-abbasi-sh/govdata-mcp"}

govdata_mcp-0.1.0/src/govdata_mcp/__init__.py

@@ -0,0 +1,7 @@
+"""govdata_mcp — unofficial read-only MCP server for govdata.de.
+
+Germany's national open-data portal govdata.de via its official CKAN API: search 163,000+ datasets, dataset details, publishers, recent datasets and resource details. No key, no login.
+"""
+from __future__ import annotations
+
+__version__ = "0.1.0"

govdata_mcp-0.1.0/src/govdata_mcp/domain.py

@@ -0,0 +1,117 @@
+"""Domain logic: shape govdata.de CKAN API responses into compact
+dicts (stdlib only).
+
+Data source: the official CKAN API of Germany's national open-data
+portal (https://www.govdata.de/ckan/api/3/action/...) — free, no key,
+no login. Actions used: package_search, package_show,
+organization_list, resource_show.
+"""
+from __future__ import annotations
+
+import json
+import urllib.parse
+
+BASE = "https://www.govdata.de/ckan/api/3/action"
+
+
+def validate_limit(limit: int) -> int:
+    if not 1 <= limit <= 50:
+        raise ValueError(f"limit out of range: {limit} (expected 1..50)")
+    return limit
+
+
+def search_url(query: str, rows: int, sort: str) -> str:
+    params = {"q": query, "rows": str(rows)}
+    if sort == "recent":
+        params["sort"] = "metadata_modified desc"
+    return BASE + "/package_search?" + urllib.parse.urlencode(params)
+
+
+def recent_url(rows: int) -> str:
+    params = {"rows": str(rows), "sort": "metadata_modified desc"}
+    return BASE + "/package_search?" + urllib.parse.urlencode(params)
+
+
+def dataset_url(dataset_id: str) -> str:
+    return BASE + "/package_show?" + urllib.parse.urlencode({"id": dataset_id})
+
+
+def publishers_url(limit: int) -> str:
+    params = {"all_fields": "true", "limit": str(limit)}
+    return BASE + "/organization_list?" + urllib.parse.urlencode(params)
+
+
+def resource_url(resource_id: str) -> str:
+    return BASE + "/resource_show?" + urllib.parse.urlencode({"id": resource_id})
+
+
+def _load(text: str, what: str) -> dict:
+    try:
+        data = json.loads(text)
+    except json.JSONDecodeError as e:
+        raise ValueError(f"invalid {what}: {e}") from e
+    if not data.get("success"):
+        err = data.get("error") or {}
+        raise ValueError(f"govdata.de error: {err.get('message') or err}")
+    return data["result"]
+
+
+def _compact_dataset(pkg: dict) -> dict:
+    org = pkg.get("organization") or {}
+    return {
+        "id": pkg.get("id") or "",
+        "name": pkg.get("name") or "",
+        "title": pkg.get("title") or "",
+        "notes": (pkg.get("notes") or "")[:400],
+        "organization": org.get("title") or org.get("name") or "",
+        "license": pkg.get("license_title") or "",
+        "num_resources": pkg.get("num_resources") or 0,
+        "tags": [t.get("name") for t in (pkg.get("tags") or [])[:5]],
+        "metadata_modified": pkg.get("metadata_modified") or "",
+        "resources": [
+            {"id": r.get("id") or "", "name": r.get("name") or "",
+             "format": r.get("format") or "", "url": r.get("url") or ""}
+            for r in (pkg.get("resources") or [])[:10]
+        ],
+    }
+
+
+def parse_search(text: str) -> dict:
+    result = _load(text, "package_search response")
+    return {
+        "total_results": result.get("count") or 0,
+        "items": [_compact_dataset(p) for p in result.get("results") or []],
+    }
+
+
+def parse_dataset(text: str) -> dict:
+    return _compact_dataset(_load(text, "package_show response"))
+
+
+def parse_publishers(text: str) -> list[dict]:
+    result = _load(text, "organization_list response")
+    orgs = [{
+        "id": o.get("id") or "",
+        "name": o.get("name") or "",
+        "title": o.get("title") or "",
+        "package_count": o.get("package_count") or 0,
+        "description": (o.get("description") or "")[:200],
+    } for o in result]
+    orgs.sort(key=lambda o: o["package_count"], reverse=True)
+    return orgs
+
+
+def parse_resource(text: str) -> dict:
+    r = _load(text, "resource_show response")
+    return {
+        "id": r.get("id") or "",
+        "name": r.get("name") or "",
+        "format": r.get("format") or "",
+        "url": r.get("url") or r.get("download_url") or "",
+        "description": (r.get("description") or "")[:400],
+        "size": r.get("size"),
+        "mimetype": r.get("mimetype") or "",
+        "created": r.get("created") or "",
+        "last_modified": r.get("last_modified") or "",
+        "package_id": r.get("package_id") or "",
+    }

govdata_mcp-0.1.0/src/govdata_mcp/http.py

@@ -0,0 +1,99 @@
+"""HTTP layer for govdata_mcp — plain httpx GETs against govdata.de.
+
+No key, no login. One retry on 429/5xx; error taxonomy (403/429/other).
+"""
+from __future__ import annotations
+
+import os
+import ssl
+import time
+
+import httpx
+
+UA = (
+    "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 "
+    "(KHTML, like Gecko) Chrome/131.0.0.0 Safari/537.36"
+)
+TIMEOUT = 25.0
+PROBE_URL = "https://www.govdata.de/ckan/api/3/action/package_search?q=berlin&rows=1"
+SOURCE = "govdata.de"
+
+
+class GovdataMcpError(RuntimeError):
+    pass
+
+
+def _verify():
+    """TLS trust store: the OS bundle when present, else httpx/certifi."""
+    for path in (
+        "/etc/ssl/certs/ca-certificates.crt",
+        ssl.get_default_verify_paths().openssl_cafile,
+    ):
+        if path and os.path.exists(path):
+            return ssl.create_default_context(cafile=path)
+    return True
+
+
+def _client() -> httpx.Client:
+    # NOTE: we read the proxy URL from the environment but never log it.
+    proxy = os.environ.get("https_proxy") or os.environ.get("HTTPS_PROXY")
+    return httpx.Client(
+        headers={"User-Agent": UA, "Accept": "application/json, application/xml, text/html, */*"},
+        timeout=TIMEOUT,
+        proxy=proxy or None,
+        trust_env=False,
+        verify=_verify(),
+    )
+
+
+def fetch_text(url: str) -> str:
+    """GET one URL and return its text. One retry on 429/5xx."""
+    last_exc: Exception | None = None
+    r: httpx.Response | None = None
+    for attempt in range(2):
+        try:
+            with _client() as c:
+                r = c.get(url)
+        except httpx.HTTPError as e:
+            last_exc = e
+            if attempt == 0:
+                time.sleep(1.0)
+                continue
+            raise GovdataMcpError(f"network error fetching {url}: {e}") from e
+        if r.status_code in (429, 500, 502, 503):
+            if attempt == 0:
+                time.sleep(1.0)
+                continue
+        break
+    assert r is not None  # loop always assigns r or raises
+    if r.status_code == 403:
+        raise GovdataMcpError(
+            f"{SOURCE} returned HTTP 403 for {url} (bot wall) — try again later"
+        )
+    if r.status_code == 429:
+        raise GovdataMcpError(
+            f"{SOURCE} rate-limited the request (HTTP 429) — try again later"
+        )
+    if r.status_code != 200:
+        raise GovdataMcpError(f"{SOURCE} returned HTTP {r.status_code} for {url}")
+    return r.text
+
+
+def health_check() -> dict:
+    """One polite probe of the source site.
+
+    Returns {"status": "ok"|"blocked"|"error", "http_status": int|None,
+    "url": ..., "detail": ...}. Used by the `<prefix>_health` tool
+    and the daily smoke CI.
+    """
+    url = PROBE_URL
+    try:
+        fetch_text(url)
+    except GovdataMcpError as e:
+        msg = str(e)
+        if "403" in msg:
+            return {"status": "blocked", "http_status": 403, "url": url, "detail": msg}
+        if "429" in msg:
+            return {"status": "blocked", "http_status": 429, "url": url, "detail": msg}
+        return {"status": "error", "http_status": None, "url": url, "detail": msg}
+    return {"status": "ok", "http_status": 200, "url": url}