diff --git a/agent_v0/server_v1/api_stream.py b/agent_v0/server_v1/api_stream.py
index 2f8347b29..8f51103e7 100644
--- a/agent_v0/server_v1/api_stream.py
+++ b/agent_v0/server_v1/api_stream.py
@@ -3322,6 +3322,197 @@ def _vlm_quick_find(
         return None
 
 
+# ---------------------------------------------------------------------------
+# Résolution Set-of-Mark : SomEngine (détection) + VLM (identification)
+# ---------------------------------------------------------------------------
+
+_som_engine_api = None  # Singleton
+
+
+def _get_som_engine_api():
+    """Singleton SomEngine pour la résolution visuelle (lazy-loaded, GPU)."""
+    global _som_engine_api
+    if _som_engine_api is None:
+        try:
+            from core.detection.som_engine import SomEngine
+            _som_engine_api = SomEngine(device="cuda")
+            logger.info("SomEngine API initialisé (lazy singleton)")
+        except Exception as e:
+            logger.warning("SomEngine API non disponible : %s", e)
+            _som_engine_api = False
+    return _som_engine_api if _som_engine_api is not False else None
+
+
+def _resolve_by_som(
+    screenshot_path: str,
+    target_spec: Dict[str, Any],
+    screen_width: int,
+    screen_height: int,
+) -> Optional[Dict[str, Any]]:
+    """Résoudre une cible UI via Set-of-Mark + VLM.
+
+    Pipeline :
+    1. SomEngine détecte tous les éléments et les numérote sur le screenshot
+    2. VLM reçoit l'image annotée + description de la cible
+    3. VLM identifie le numéro du mark → coordonnées précises
+
+    Avantages vs VLM direct :
+    - Le VLM n'a qu'à identifier (son point fort), pas localiser
+    - Les coordonnées viennent de SomEngine (pixel-perfect)
+    - Question simple "quel numéro ?" → réponse simple
+
+    Args:
+        screenshot_path: Chemin du screenshot actuel
+        target_spec: Spécification de la cible (vlm_description, som_element, etc.)
+        screen_width: Largeur écran en pixels
+        screen_height: Hauteur écran en pixels
+
+    Returns:
+        Dict avec resolved=True et coordonnées, ou None si indisponible.
+    """
+    engine = _get_som_engine_api()
+    if engine is None:
+        return None
+
+    client = _get_vlm_client()
+    if client is None:
+        return None
+
+    t0 = time.time()
+
+    # ── 1. Lancer SomEngine sur le screenshot actuel ──
+    try:
+        from PIL import Image as PILImage
+        img = PILImage.open(screenshot_path).convert("RGB")
+        som_result = engine.analyze(img)
+    except Exception as e:
+        logger.warning("SoM resolve : erreur analyse — %s", e)
+        return None
+
+    if not som_result.elements:
+        logger.info("SoM resolve : 0 éléments détectés")
+        return None
+
+    # ── 2. Construire la description de la cible ──
+    som_element = target_spec.get("som_element", {})
+    vlm_description = target_spec.get("vlm_description", "")
+    anchor_label = som_element.get("label", "")
+
+    # Construire un prompt riche
+    target_parts = []
+    if anchor_label:
+        target_parts.append(f"texte '{anchor_label}'")
+    if vlm_description:
+        target_parts.append(vlm_description)
+    if not target_parts:
+        # Sans description, SoM resolve ne peut pas fonctionner
+        logger.debug("SoM resolve : pas de description pour identifier l'élément")
+        return None
+
+    target_desc = ", ".join(target_parts)
+
+    # ── 3. Sauvegarder l'image annotée SoM temporairement ──
+    import tempfile
+    try:
+        with tempfile.NamedTemporaryFile(suffix=".jpg", delete=False) as tmp:
+            som_result.som_image.save(tmp, format="JPEG", quality=85)
+            som_img_path = tmp.name
+    except Exception as e:
+        logger.warning("SoM resolve : erreur sauvegarde image annotée — %s", e)
+        return None
+
+    # ── 4. VLM : identifier le numéro du mark ──
+    # Lister les éléments avec leur numéro pour aider le VLM
+    elements_list = "\n".join(
+        f"  #{e.id}: '{e.label}' ({e.source})"
+        for e in som_result.elements[:50]  # Limiter à 50 éléments
+        if e.label
+    )
+
+    prompt = (
+        "This screenshot has numbered marks (red badges) on each UI element.\n\n"
+        f"I'm looking for this element: {target_desc}\n\n"
+    )
+    if elements_list:
+        prompt += f"Detected elements:\n{elements_list}\n\n"
+    prompt += (
+        "Which mark number corresponds to this element?\n"
+        'Return ONLY a JSON object: {"mark_id": N, "confidence": 0.XX}\n'
+        "If not found, return: {\"mark_id\": null, \"confidence\": 0.0}"
+    )
+
+    system_prompt = "You are a UI element identifier. Look at numbered marks on the screenshot. Output raw JSON only."
+
+    try:
+        result = client.generate(
+            prompt=prompt,
+            image_path=som_img_path,
+            system_prompt=system_prompt,
+            temperature=0.1,
+            max_tokens=100,
+            force_json=False,
+        )
+    except Exception as e:
+        logger.warning("SoM resolve : erreur VLM — %s", e)
+        return None
+    finally:
+        import os
+        try:
+            os.unlink(som_img_path)
+        except OSError:
+            pass
+
+    elapsed = time.time() - t0
+
+    if not result.get("success"):
+        logger.info("SoM resolve : VLM échoué (%.1fs)", elapsed)
+        return None
+
+    # ── 5. Parser la réponse et retourner les coordonnées ──
+    response_text = result.get("response", "").strip()
+    parsed = client._extract_json_from_response(response_text)
+    if parsed is None:
+        logger.info("SoM resolve : réponse non-JSON (%.1fs) — %.80s", elapsed, response_text)
+        return None
+
+    mark_id = parsed.get("mark_id")
+    confidence = float(parsed.get("confidence", 0.0))
+
+    if mark_id is None or confidence < 0.3:
+        logger.info(
+            "SoM resolve : mark non trouvé ou confiance trop basse (mark=%s, conf=%.2f, %.1fs)",
+            mark_id, confidence, elapsed,
+        )
+        return None
+
+    mark_id = int(mark_id)
+    elem = som_result.get_element_by_id(mark_id)
+    if elem is None:
+        logger.warning("SoM resolve : mark #%d inexistant (%.1fs)", mark_id, elapsed)
+        return None
+
+    cx_norm, cy_norm = elem.center_norm
+    logger.info(
+        "SoM resolve OK : mark #%d '%s' → (%.4f, %.4f) conf=%.2f en %.1fs (%d éléments)",
+        mark_id, elem.label, cx_norm, cy_norm, confidence, elapsed, len(som_result.elements),
+    )
+
+    return {
+        "resolved": True,
+        "method": "som_vlm",
+        "x_pct": round(cx_norm, 6),
+        "y_pct": round(cy_norm, 6),
+        "matched_element": {
+            "label": elem.label or f"mark #{mark_id}",
+            "type": elem.source,
+            "role": "som_identified",
+            "confidence": confidence,
+            "som_id": mark_id,
+        },
+        "score": confidence,
+    }
+
+
 def _resolve_target_sync(
     screenshot_path: str,
     target_spec: Dict[str, Any],
@@ -3336,6 +3527,7 @@ def _resolve_target_sync(
     Hiérarchie de résolution (strict_mode=True, replay sessions) — VLM-FIRST :
     1. VLM Quick Find (~3-8s) — compréhension sémantique de l'écran, multi-image
        (screenshot + crop de référence + description riche)
+    1.5. SoM + VLM (~5-15s) — SomEngine numérote les éléments, VLM identifie le bon
     2. Template matching OpenCV (~100ms) — fallback pixel, seuil STRICT 0.90
     3. resolved=False → STOP le replay
 
@@ -3394,6 +3586,30 @@ def _resolve_target_sync(
                     vlm_description[:60] if vlm_description else "(anchor)",
                 )
 
+        # ---------------------------------------------------------------
+        # Étape 1.5 : SoM + VLM (Set-of-Mark + identification)
+        # SomEngine numérote les éléments, VLM identifie le bon numéro.
+        # Plus fiable que le VLM direct car le VLM n'a qu'à identifier,
+        # pas localiser — et les coordonnées sont pixel-perfect.
+        # ---------------------------------------------------------------
+        som_element = target_spec.get("som_element", {})
+        if som_element or vlm_description:
+            som_result = _resolve_by_som(
+                screenshot_path=screenshot_path,
+                target_spec=target_spec,
+                screen_width=screen_width,
+                screen_height=screen_height,
+            )
+            if som_result and som_result.get("resolved"):
+                logger.info(
+                    "Strict resolve SoM+VLM : OK (score=%.2f, mark=#%s)",
+                    som_result.get("score", 0),
+                    som_result.get("matched_element", {}).get("som_id", "?"),
+                )
+                return som_result
+            else:
+                logger.info("Strict resolve SoM+VLM : échoué, passage template matching")
+
         # ---------------------------------------------------------------
         # Étape 2 : Template matching (fallback pixel) — seuil STRICT 0.90
         # ---------------------------------------------------------------
diff --git a/agent_v0/server_v1/stream_processor.py b/agent_v0/server_v1/stream_processor.py
index 379293391..6f1afd105 100644
--- a/agent_v0/server_v1/stream_processor.py
+++ b/agent_v0/server_v1/stream_processor.py
@@ -427,6 +427,111 @@ def _needs_post_wait(action: dict) -> int:
     return 0
 
 
+# ---------------------------------------------------------------------------
+# SomEngine — enrichissement Set-of-Mark des clics pendant le build_replay
+# ---------------------------------------------------------------------------
+
+_som_engine = None  # Singleton, chargé à la demande
+
+
+def _get_som_engine():
+    """Singleton SomEngine (lazy-loaded, GPU)."""
+    global _som_engine
+    if _som_engine is None:
+        try:
+            from core.detection.som_engine import SomEngine
+            _som_engine = SomEngine(device="cuda")
+            logger.info("SomEngine initialisé (lazy singleton)")
+        except Exception as e:
+            logger.warning("SomEngine non disponible : %s", e)
+            _som_engine = False  # Marqueur "indisponible"
+    return _som_engine if _som_engine is not False else None
+
+
+def _som_identify_clicked_element(
+    event_data: dict,
+    session_dir: Optional[Path],
+    screen_w: int,
+    screen_h: int,
+) -> Optional[dict]:
+    """Identifier l'élément UI cliqué via SomEngine (YOLO + docTR).
+
+    Charge le full screenshot de l'événement, lance SomEngine pour détecter
+    tous les éléments, puis identifie celui qui se trouve sous le clic.
+
+    Returns:
+        Dict avec id, label, source, bbox_norm, center_norm, confidence
+        ou None si SomEngine indisponible ou élément non trouvé.
+    """
+    engine = _get_som_engine()
+    if engine is None:
+        return None
+
+    if not session_dir:
+        return None
+
+    shots_dir = session_dir / "shots"
+    if not shots_dir.is_dir():
+        return None
+
+    # Trouver le full screenshot
+    screenshot_id = event_data.get("screenshot_id", "")
+    if not screenshot_id:
+        return None
+
+    full_path = shots_dir / f"{screenshot_id}_full.png"
+    if not full_path.is_file():
+        # Fallback : essayer sans le suffixe _full
+        full_path = shots_dir / f"{screenshot_id}.png"
+        if not full_path.is_file():
+            return None
+
+    try:
+        from PIL import Image
+        img = Image.open(full_path).convert("RGB")
+    except Exception as e:
+        logger.debug("SoM: impossible de charger %s : %s", full_path, e)
+        return None
+
+    # Lancer SomEngine
+    try:
+        result = engine.analyze(img)
+    except Exception as e:
+        logger.warning("SoM: erreur d'analyse : %s", e)
+        return None
+
+    if not result.elements:
+        return None
+
+    # Trouver l'élément cliqué
+    pos = event_data.get("pos", [])
+    if not pos or len(pos) < 2:
+        return None
+
+    click_x, click_y = int(pos[0]), int(pos[1])
+    elem = result.find_element_at(click_x, click_y, margin=30)
+    if elem is None:
+        logger.debug(
+            "SoM: aucun élément trouvé au clic (%d, %d) parmi %d éléments",
+            click_x, click_y, len(result.elements),
+        )
+        return None
+
+    logger.info(
+        "SoM: clic (%d,%d) → élément #%d '%s' (source=%s, conf=%.2f)",
+        click_x, click_y, elem.id, elem.label, elem.source, elem.confidence,
+    )
+    return {
+        "id": elem.id,
+        "label": elem.label,
+        "source": elem.source,
+        "bbox_norm": list(elem.bbox_norm),
+        "center_norm": list(elem.center_norm),
+        "confidence": elem.confidence,
+        "element_count": len(result.elements),
+    }
+
+
 def _load_crop_for_event(
     event_data: dict,
     session_dir: Optional[Path],
@@ -919,6 +1024,18 @@ def build_replay_from_raw_events(
                     # Sinon le template matching texte cherche "13071967.txt – Bloc-notes"
                     # sur l'écran et clique sur la barre de titre au lieu du bon élément.
 
+                    # ── SomEngine : identifier l'élément cliqué ──
+                    som_elem = _som_identify_clicked_element(
+                        evt, session_dir_path, screen_w, screen_h,
+                    )
+                    if som_elem:
+                        action["target_spec"]["som_element"] = som_elem
+                        # Enrichir la description VLM avec le label SoM
+                        if som_elem.get("label") and not vision_info.get("text"):
+                            action["target_spec"]["vlm_description"] += (
+                                f", le texte de l'élément est '{som_elem['label']}'"
+                            )
+
         elif evt_type == "text_input":
             text = evt.get("text", "")
             if not text:
diff --git a/tests/unit/conftest.py b/tests/unit/conftest.py
new file mode 100644
index 000000000..659a04127
--- /dev/null
+++ b/tests/unit/conftest.py
@@ -0,0 +1,26 @@
+"""Conftest pour les tests unitaires.
+
+Force le bon chemin agent_v0 (rpa_vision_v3) pour éviter les conflits
+avec ~/ai/agent_v0 (standalone).
+"""
+import sys
+from pathlib import Path
+
+ROOT = str(Path(__file__).resolve().parents[2])
+
+if ROOT in sys.path:
+    sys.path.remove(ROOT)
+sys.path.insert(0, ROOT)
+
+# Si agent_v0 est déjà chargé depuis le mauvais chemin, le remplacer
+_agent_mod = sys.modules.get("agent_v0")
+if _agent_mod and not getattr(_agent_mod, "__file__", "").startswith(ROOT):
+    to_remove = [k for k in sys.modules if k == "agent_v0" or k.startswith("agent_v0.")]
+    for k in to_remove:
+        del sys.modules[k]
+
+# Pré-importer le bon agent_v0.server_v1
+try:
+    import agent_v0.server_v1  # noqa: F401
+except ImportError:
+    pass
diff --git a/tests/unit/test_som_integration.py b/tests/unit/test_som_integration.py
new file mode 100644
index 000000000..ac35c4687
--- /dev/null
+++ b/tests/unit/test_som_integration.py
@@ -0,0 +1,301 @@
+"""Tests unitaires pour l'intégration SomEngine dans build_replay et resolve_target.
+
+Vérifie :
+- Phase 1 : _som_identify_clicked_element enrichit target_spec avec som_element
+- Phase 2 : _resolve_by_som utilise SomEngine + VLM pour résoudre une cible
+- Fallbacks gracieux quand SomEngine ou VLM indisponible
+"""
+
+from pathlib import Path
+from unittest.mock import MagicMock, patch
+
+import pytest
+
+
+# ── Phase 1 : Enrichissement build_replay ──
+
+
+class TestSomIdentifyClickedElement:
+    """Tests pour _som_identify_clicked_element (Phase 1)."""
+
+    def test_returns_none_when_engine_unavailable(self):
+        """Si SomEngine n'est pas disponible, retourne None sans erreur."""
+        from agent_v0.server_v1.stream_processor import _som_identify_clicked_element
+
+        with patch(
+            "agent_v0.server_v1.stream_processor._get_som_engine",
+            return_value=None,
+        ):
+            result = _som_identify_clicked_element(
+                {"screenshot_id": "shot_0001", "pos": [500, 300]},
+                Path("/fake/dir"),
+                1920, 1080,
+            )
+            assert result is None
+
+    def test_returns_none_when_no_session_dir(self):
+        """Sans session_dir, retourne None."""
+        from agent_v0.server_v1.stream_processor import _som_identify_clicked_element
+
+        result = _som_identify_clicked_element(
+            {"screenshot_id": "shot_0001", "pos": [500, 300]},
+            None, 1920, 1080,
+        )
+        assert result is None
+
+    def test_returns_none_when_no_screenshot_id(self):
+        """Sans screenshot_id, retourne None."""
+        from agent_v0.server_v1.stream_processor import _som_identify_clicked_element
+
+        result = _som_identify_clicked_element(
+            {"pos": [500, 300]},
+            Path("/fake/dir"),
+            1920, 1080,
+        )
+        assert result is None
+
+    def test_returns_element_when_found(self, tmp_path):
+        """Quand SomEngine trouve un élément sous le clic, retourne ses infos."""
+        from core.detection.som_engine import SomElement, SomResult
+        from agent_v0.server_v1.stream_processor import _som_identify_clicked_element
+
+        # Créer un faux screenshot
+        shots_dir = tmp_path / "shots"
+        shots_dir.mkdir()
+        from PIL import Image
+        img = Image.new("RGB", (1920, 1080), color="white")
+        img.save(shots_dir / "shot_0001_full.png")
+
+        # Mock SomEngine
+        mock_elem = SomElement(
+            id=5,
+            bbox=(480, 280, 520, 320),
+            bbox_norm=(0.25, 0.259, 0.271, 0.296),
+            center=(500, 300),
+            center_norm=(0.2604, 0.2778),
+            source="yolo",
+            label="Enregistrer",
+            confidence=0.92,
+        )
+        mock_result = SomResult(
+            elements=[mock_elem],
+            width=1920,
+            height=1080,
+        )
+        mock_engine = MagicMock()
+        mock_engine.analyze.return_value = mock_result
+
+        with patch(
+            "agent_v0.server_v1.stream_processor._get_som_engine",
+            return_value=mock_engine,
+        ):
+            result = _som_identify_clicked_element(
+                {"screenshot_id": "shot_0001", "pos": [500, 300]},
+                tmp_path, 1920, 1080,
+            )
+
+        assert result is not None
+        assert result["id"] == 5
+        assert result["label"] == "Enregistrer"
+        assert result["source"] == "yolo"
+        assert result["confidence"] == 0.92
+        assert result["element_count"] == 1
+
+    def test_returns_none_when_no_element_at_click(self, tmp_path):
+        """Quand aucun élément n'est sous le clic, retourne None."""
+        from core.detection.som_engine import SomResult
+        from agent_v0.server_v1.stream_processor import _som_identify_clicked_element
+
+        shots_dir = tmp_path / "shots"
+        shots_dir.mkdir()
+        from PIL import Image
+        img = Image.new("RGB", (1920, 1080), color="white")
+        img.save(shots_dir / "shot_0001_full.png")
+
+        # Résultat avec des éléments mais pas au point du clic
+        mock_result = SomResult(elements=[], width=1920, height=1080)
+        mock_engine = MagicMock()
+        mock_engine.analyze.return_value = mock_result
+
+        with patch(
+            "agent_v0.server_v1.stream_processor._get_som_engine",
+            return_value=mock_engine,
+        ):
+            result = _som_identify_clicked_element(
+                {"screenshot_id": "shot_0001", "pos": [500, 300]},
+                tmp_path, 1920, 1080,
+            )
+
+        assert result is None
+
+
+# ── Phase 2 : Résolution SoM + VLM ──
+
+
+class TestResolveBySom:
+    """Tests pour _resolve_by_som (Phase 2)."""
+
+    def test_returns_none_when_engine_unavailable(self):
+        """Sans SomEngine, retourne None."""
+        from agent_v0.server_v1.api_stream import _resolve_by_som
+
+        with patch(
+            "agent_v0.server_v1.api_stream._get_som_engine_api",
+            return_value=None,
+        ):
+            result = _resolve_by_som(
+                "/fake/path.jpg",
+                {"vlm_description": "un bouton"},
+                1920, 1080,
+            )
+            assert result is None
+
+    def test_returns_none_when_vlm_unavailable(self):
+        """Sans VLM, retourne None."""
+        from agent_v0.server_v1.api_stream import _resolve_by_som
+
+        mock_engine = MagicMock()
+
+        with patch(
+            "agent_v0.server_v1.api_stream._get_som_engine_api",
+            return_value=mock_engine,
+        ), patch(
+            "agent_v0.server_v1.api_stream._get_vlm_client",
+            return_value=None,
+        ):
+            result = _resolve_by_som(
+                "/fake/path.jpg",
+                {"vlm_description": "un bouton"},
+                1920, 1080,
+            )
+            assert result is None
+
+    def test_returns_none_without_description(self):
+        """Sans description ni som_element, retourne None."""
+        from agent_v0.server_v1.api_stream import _resolve_by_som
+
+        mock_engine = MagicMock()
+        mock_client = MagicMock()
+
+        with patch(
+            "agent_v0.server_v1.api_stream._get_som_engine_api",
+            return_value=mock_engine,
+        ), patch(
+            "agent_v0.server_v1.api_stream._get_vlm_client",
+            return_value=mock_client,
+        ):
+            result = _resolve_by_som(
+                "/fake/path.jpg",
+                {},  # Pas de description
+                1920, 1080,
+            )
+            assert result is None
+
+    def test_resolve_success(self, tmp_path):
+        """Résolution réussie : SomEngine détecte, VLM identifie le mark."""
+        from core.detection.som_engine import SomElement, SomResult
+        from agent_v0.server_v1.api_stream import _resolve_by_som
+
+        # Créer un faux screenshot
+        from PIL import Image
+        img = Image.new("RGB", (1920, 1080), color="white")
+        screenshot_path = str(tmp_path / "screen.jpg")
+        img.save(screenshot_path)
+
+        # Mock SomEngine
+        mock_elem = SomElement(
+            id=9,
+            bbox=(960, 540, 1000, 570),
+            bbox_norm=(0.5, 0.5, 0.521, 0.528),
+            center=(980, 555),
+            center_norm=(0.5104, 0.5139),
+            source="ocr",
+            label="Ouvrir",
+            confidence=0.88,
+        )
+        mock_result = SomResult(
+            elements=[mock_elem],
+            som_image=img.copy(),
+            som_image_b64="fake_b64",
+            width=1920,
+            height=1080,
+        )
+        mock_engine = MagicMock()
+        mock_engine.analyze.return_value = mock_result
+
+        # Mock VLM client
+        mock_client = MagicMock()
+        mock_client.generate.return_value = {
+            "success": True,
+            "response": '{"mark_id": 9, "confidence": 0.95}',
+        }
+        mock_client._extract_json_from_response.return_value = {
+            "mark_id": 9,
+            "confidence": 0.95,
+        }
+
+        with patch(
+            "agent_v0.server_v1.api_stream._get_som_engine_api",
+            return_value=mock_engine,
+        ), patch(
+            "agent_v0.server_v1.api_stream._get_vlm_client",
+            return_value=mock_client,
+        ):
+            result = _resolve_by_som(
+                screenshot_path,
+                {
+                    "vlm_description": "le bouton Ouvrir",
+                    "som_element": {"id": 9, "label": "Ouvrir"},
+                },
+                1920, 1080,
+            )
+
+        assert result is not None
+        assert result["resolved"] is True
+        assert result["method"] == "som_vlm"
+        assert abs(result["x_pct"] - 0.5104) < 0.001
+        assert abs(result["y_pct"] - 0.5139) < 0.001
+        assert result["matched_element"]["som_id"] == 9
+
+    def test_resolve_vlm_low_confidence(self, tmp_path):
+        """VLM retourne une confiance trop basse → None."""
+        from core.detection.som_engine import SomResult
+        from agent_v0.server_v1.api_stream import _resolve_by_som
+
+        from PIL import Image
+        img = Image.new("RGB", (1920, 1080), color="white")
+        screenshot_path = str(tmp_path / "screen.jpg")
+        img.save(screenshot_path)
+
+        mock_result = SomResult(
+            elements=[MagicMock(id=1, label="test", source="ocr")],
+            som_image=img.copy(),
+            width=1920, height=1080,
+        )
+        mock_engine = MagicMock()
+        mock_engine.analyze.return_value = mock_result
+
+        mock_client = MagicMock()
+        mock_client.generate.return_value = {
+            "success": True,
+            "response": '{"mark_id": 1, "confidence": 0.1}',
+        }
+        mock_client._extract_json_from_response.return_value = {
+            "mark_id": 1,
+            "confidence": 0.1,
+        }
+
+        with patch(
+            "agent_v0.server_v1.api_stream._get_som_engine_api",
+            return_value=mock_engine,
+        ), patch(
+            "agent_v0.server_v1.api_stream._get_vlm_client",
+            return_value=mock_client,
+        ):
+            result = _resolve_by_som(
+                screenshot_path,
+                {"vlm_description": "un bouton"},
+                1920, 1080,
+            )
+
+        assert result is None