feat(v3.12.0): media pipeline, AI image generation, capability probe, companion v2.9.0

Three-month batch sync from internal repo (~80 commits) covering Tracks F.5a, F.7e, F.8, F.17, F.18, F.X.

WordPress media pipeline
- Pillow-based optimization, AI image generation (OpenAI / Stability / Replicate / Google Nano Banana / OpenRouter), chunked + resumable uploads, bulk delete/reassign, idempotent retries.

Capability discovery (F.7e)
- Per-site credential probe + adapters for WordPress / WooCommerce / Gitea, tier-fit unions granted ∪ roles, capability badge UI with HTMX partial re-check, install hint in every companion-unreachable error.

Companion plugin overhaul
- Renamed wordpress-plugin/airano-mcp-seo-bridge → wordpress-plugin/airano-mcp-bridge.
- Eight new endpoints: /capabilities, /bulk-meta, /export, /cache-purge, /transient-flush, /site-health, /audit-hook, /upload-and-attach.
- wp.org Plugin Check pass: i18n, WP_Filesystem, scheme allowlist on audit-hook URL.

Other
- Gitea ergonomics (F.17): batch files, tree, search, compare, releases, fork.
- Opportunistic bcrypt upgrade for legacy SHA-256 admin keys (F.8).
- n8n refactor: structured errors, capability probe, missing tools backfilled.
- Idempotency-Key dedup for AI media upload retries; WP client fast-fails on unreachable sites.

Docs
- README + CLAUDE.md drop the fixed "633 tools" claim. The total grows with each release; per-plugin approximations + dashboard-surfaced counts replace it.
- Tools/Tests badges removed in favour of "Plugins: 10".

Deployment
- PyPI mirror chain, optional BUILD_HTTP_PROXY, Alpine→Yandex apk mirror, Debian-slim Plan-B Dockerfile, mirror.gcr.io variant.

CI
- Black + Ruff clean on Python 3.12; pytest tests/ green.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

tests/plugins/wordpress/test_openrouter_model_discovery.py

@@ -0,0 +1,192 @@
+"""F.X.fix #12 — OpenRouter /v1/models discovery + 1h cache."""
+
+from __future__ import annotations
+
+from unittest.mock import AsyncMock, patch
+
+import pytest
+
+from plugins.ai_image.providers import openrouter as or_mod
+from plugins.ai_image.providers.openrouter import OpenRouterProvider
+
+
+def _fake_resp(*, status, json_data=None, headers=None):
+    resp = AsyncMock()
+    resp.status = status
+    resp.json = AsyncMock(return_value=json_data or {})
+    resp.headers = headers or {}
+    return resp
+
+
+def _mock_session(responses):
+    iterator = iter(responses)
+
+    def _request(*_a, **_kw):
+        resp = next(iterator)
+        return AsyncMock(
+            __aenter__=AsyncMock(return_value=resp),
+            __aexit__=AsyncMock(return_value=False),
+        )
+
+    sess = AsyncMock()
+    sess.get = _request
+    sess.post = _request
+    return sess
+
+
+def _reset_cache():
+    or_mod._models_cache["fetched_at"] = 0.0
+    or_mod._models_cache["payload"] = None
+
+
+@pytest.fixture(autouse=True)
+def reset_cache():
+    _reset_cache()
+    yield
+    _reset_cache()
+
+
+CATALOG_PAYLOAD = {
+    "data": [
+        {
+            "id": "google/gemini-2.5-flash-image",
+            "name": "Gemini 2.5 Flash Image",
+            "description": "Google's GA image model",
+            "context_length": 1000000,
+            "architecture": {
+                "input_modalities": ["text", "image"],
+                "output_modalities": ["image", "text"],
+                "modality": "text+image->image+text",
+            },
+        },
+        {
+            "id": "google/gemini-2.5-flash-image-preview",
+            "name": "Gemini 2.5 Flash Image (Preview)",
+            "architecture": {
+                "output_modalities": ["image"],
+            },
+        },
+        {
+            "id": "openai/gpt-4o",
+            "name": "GPT-4o",
+            "architecture": {
+                "input_modalities": ["text", "image"],
+                "output_modalities": ["text"],
+            },
+        },
+        {
+            "id": "openai/dall-e-3",
+            "name": "DALL-E 3",
+            "architecture": {"output_modalities": ["image"]},
+        },
+        {
+            "id": "meta/llama-3",
+            "name": "Llama 3",
+            "description": "chat only",
+            "architecture": {"output_modalities": ["text"]},
+        },
+    ]
+}
+
+
+class TestListImageModels:
+    @pytest.mark.asyncio
+    async def test_filters_to_image_output_only(self):
+        provider = OpenRouterProvider()
+        resp = _fake_resp(status=200, json_data=CATALOG_PAYLOAD)
+        sess = _mock_session([resp])
+        with patch("plugins.ai_image.providers.openrouter.aiohttp.ClientSession") as cls:
+            cls.return_value = AsyncMock(
+                __aenter__=AsyncMock(return_value=sess),
+                __aexit__=AsyncMock(return_value=False),
+            )
+            models = await provider.list_image_models()
+        ids = {m["id"] for m in models}
+        # Kept — image output
+        assert "google/gemini-2.5-flash-image" in ids
+        assert "openai/dall-e-3" in ids
+        # Dropped — text-only
+        assert "openai/gpt-4o" not in ids
+        assert "meta/llama-3" not in ids
+        # Dropped — deprecated alias (we refuse to surface it)
+        assert "google/gemini-2.5-flash-image-preview" not in ids
+
+    @pytest.mark.asyncio
+    async def test_shape_includes_pricing_and_modalities(self):
+        provider = OpenRouterProvider()
+        resp = _fake_resp(status=200, json_data=CATALOG_PAYLOAD)
+        sess = _mock_session([resp])
+        with patch("plugins.ai_image.providers.openrouter.aiohttp.ClientSession") as cls:
+            cls.return_value = AsyncMock(
+                __aenter__=AsyncMock(return_value=sess),
+                __aexit__=AsyncMock(return_value=False),
+            )
+            models = await provider.list_image_models()
+        gemini = next(m for m in models if m["id"] == "google/gemini-2.5-flash-image")
+        assert gemini["name"]
+        assert "image" in gemini["output_modalities"]
+        # F.X.fix #11 integration — catalog exposes price per image for
+        # UI display alongside the model list.
+        assert gemini["price_per_image_usd"] is not None
+
+    @pytest.mark.asyncio
+    async def test_cache_hits_second_call(self):
+        provider = OpenRouterProvider()
+        resp = _fake_resp(status=200, json_data=CATALOG_PAYLOAD)
+        sess = _mock_session([resp])  # only ONE response
+        with patch("plugins.ai_image.providers.openrouter.aiohttp.ClientSession") as cls:
+            cls.return_value = AsyncMock(
+                __aenter__=AsyncMock(return_value=sess),
+                __aexit__=AsyncMock(return_value=False),
+            )
+            first = await provider.list_image_models()
+            second = await provider.list_image_models()  # cache hit
+        assert first == second
+        # Second call should not have tried a second HTTP request (mock
+        # session would raise StopIteration if it did).
+
+    @pytest.mark.asyncio
+    async def test_force_bypasses_cache(self):
+        provider = OpenRouterProvider()
+        resp1 = _fake_resp(status=200, json_data=CATALOG_PAYLOAD)
+        resp2 = _fake_resp(status=200, json_data={"data": []})
+        sess = _mock_session([resp1, resp2])
+        with patch("plugins.ai_image.providers.openrouter.aiohttp.ClientSession") as cls:
+            cls.return_value = AsyncMock(
+                __aenter__=AsyncMock(return_value=sess),
+                __aexit__=AsyncMock(return_value=False),
+            )
+            first = await provider.list_image_models()
+            second = await provider.list_image_models(force=True)
+        assert len(first) > 0
+        assert second == []
+
+    @pytest.mark.asyncio
+    async def test_upstream_error_returns_previous_cache(self):
+        provider = OpenRouterProvider()
+        ok = _fake_resp(status=200, json_data=CATALOG_PAYLOAD)
+        err = _fake_resp(status=500, json_data={"error": "boom"})
+        sess = _mock_session([ok, err])
+        with patch("plugins.ai_image.providers.openrouter.aiohttp.ClientSession") as cls:
+            cls.return_value = AsyncMock(
+                __aenter__=AsyncMock(return_value=sess),
+                __aexit__=AsyncMock(return_value=False),
+            )
+            first = await provider.list_image_models()
+            second = await provider.list_image_models(force=True)
+        # First populated cache, second upstream failed — we serve the
+        # cached snapshot rather than an empty list.
+        assert second == first
+
+    @pytest.mark.asyncio
+    async def test_upstream_error_with_empty_cache_returns_empty(self):
+        provider = OpenRouterProvider()
+        err = _fake_resp(status=500, json_data={})
+        sess = _mock_session([err])
+        with patch("plugins.ai_image.providers.openrouter.aiohttp.ClientSession") as cls:
+            cls.return_value = AsyncMock(
+                __aenter__=AsyncMock(return_value=sess),
+                __aexit__=AsyncMock(return_value=False),
+            )
+            models = await provider.list_image_models()
+        assert models == []