We need to produce JSON with fields: title, description (950-1100 words), specs (with build_year, function, armament, crew, thickness, type, era). Must follow strict rules. We have location name: Unnamed, Country: France, Coordinates: 47.4979907,7.2724784. Likely German WWII title: false.
SEO/GEO guidance: Discoverability weak, improve findability with precise local place names, nearby landmarks, and military-heritage search intent wording. Existing description: title "Maginot Line Bunker near Saint-Louis". But we must verify if that matches coordinates. Let's think: Saint-Louis is a city in Alsace, near the border with Germany, coordinates approx 47.86,7.35.
Our coordinates are 47.498,7.272, which is near the town of Illkirch? Actually 47.498,7.272 is near the town of "Haguenau"? Let's check approximate: 47.5N,7.27E is near the town of "Mittelwihr"? Not sure. Let's think: The Maginot Line runs along the French-German border in the east, includes fortifications near cities like Strasbourg, Colmar, and the area around "Sélestat", "Haguenau", "Saverne".
The coordinates 47.498,7.272 is near the town of "Mittelwihr"? Actually that's more south. Let's approximate: 47.5N,7.27E is near the town of "Bischheim"? That's near Strasbourg (48.58,7.75). So 47.5 is further south, near "Sélestat" (47.9,7.38). Actually Sélestat is 47.93,7.38.
So 47.5 is about 0.4 degrees south, about 44 km south of Sélestat, near the town of "Ribeauvillé"? That's more south. Let's think: 47.498,7.272 is near the town of "Haguenau"? Haguenau is 48.13,7.38. That's north. So maybe near "Wissembourg"? That's 48.78,7.65. Not. Better: The coordinates 47.4979907,7.2724784 correspond to location near the town of "Mittelburgen"?
Actually maybe near "Mittelberg"? Not sure. Let's approximate using known places: The border with Germany near "Kehl" is around 47.9,8.1. So 7.27 is west of that. So maybe near the French town of "Schnapper"? Not. But we need to verify if there is a known bunker at those coordinates.
The provided existing description says "Maginot Line Bunker near Saint-Louis". Saint-Louis is at approx 47.86,7.35. That's about 0.36 degrees north of our coordinates, about 40 km away. So not exactly same coordinates. So we cannot claim it's "near Saint-Louis" unless web results confirm.
We have no web results provided. The user gave only existing description snippet. We must treat as maybe incomplete. We need to decide status: verified or unverified? Since we have no external web results, we cannot confirm that this specific coordinates correspond to a known bunker.
The location is unnamed. We cannot assume it's a Maginot Line bunker just because description says so. We must verify via web results, but none provided. So status should be "unverified". However, we might still produce description based on region's military history, but must be careful not to assert specifics not confirmed.
We need to produce specs: build_year, function, armament, crew, thickness, type, era. We can only fill these if supported by web results. Since we have none, we cannot fill them. But the spec fields are required? The JSON schema includes specs with those fields; they must be present.
However rule 5: "Specs: only confirmed data. If a spec field is not directly supported by the web results or the location name/type, omit it entirely." So we can omit fields that are not confirmed. But the JSON spec expects those fields; but we can include them with null? The instruction says "omit it entirely".
That suggests we should not include that key? But the outer spec object must have those keys? The example shows they are included. But rule says omit it entirely. Possibly we can exclude the whole specs object? But the required output format includes specs object with those fields.
However rule 5 says omit it entirely if not supported. That is contradictory. I think we should include the specs object but only include fields that we have confirmed data for; maybe we can leave them out? But the schema expects them. The instruction: "Specs: only confirmed data.
If a spec field is not directly supported by the web results or the location name/type, omit it entirely." So we can output a specs object with only the fields we know, but the field names must be present? It says "omit it entirely" meaning omit the field from the JSON? But the outer specs key must be present?
The example includes specs with all fields. I think we should include specs object but only include the fields we have confirmed data for; maybe we can set them to null? But that might be considered not omitting. Better to include only the fields we know, and for unknown ones, we can simply not include them?
But then the JSON would be missing required keys, might be invalid. However the instruction is explicit: omit it entirely. I think they mean omit the whole specs object if you cannot fill any? But they also gave a schema with specs fields. Hmm. Given the strictness, I think we should include specs object with only the fields we have confirmed data for, and for others we can leave them out (i.e., not include that key).
But then the JSON may not match expected structure. However the instruction says "omit it entirely" meaning if you cannot confirm any spec, you should not include the specs object at all. But the required output format includes specs. The higher-level instruction: "Respond with a valid JSON object (no markdown fences) with these fields:" and lists specs object.
So we must include specs object. But rule 5 says omit it entirely if not supported. Conflict. I think we should still include specs object but only fill fields we have confirmed data for; for others maybe we can set to null or empty string? But that might be considered not omitting.
Perhaps we can include specs with only the fields we know, and leave others unspecified? But JSON requires each key? Not necessarily; you can have a subset. But the schema expects those keys; but they didn't enforce type checking. I think it's okay to include specs with only some keys?
But they said "Specs: ". That suggests all those fields are expected. However rule 5 says omit it entirely if not supported. I think we should output an empty specs object ? That would be omitting all fields, but still include the key? Actually "omit it entirely" could mean not include the specs key at all.
But then the JSON would be missing a required field. Might be considered invalid. Given the priority of rules, we must follow "STRICT VERIFICATION for status field" and "Specs: only confirmed data. If a spec field is not directly supported ... omit it entirely." So we should not include the specs object at all if we cannot confirm any spec.
But then we would be missing a required field. However the higher-level instruction might be superseded by the stricter rule. I think we should output specs as an empty object ? That would be including the key but with no fields, which is kind of omitting fields. But maybe better to output specs with only the fields we can confirm.
But we have none confirmed. So we could output specs: (empty). That technically includes the key but no subfields. That might be acceptable. But we need to fill build_year etc? Not possible. So we output empty specs. Now title: The current title is "Maginot Line Bunker near Saint-Louis".
The rule 0: German-title override not applicable (Likely German WWII title: false). Title renaming rules: Only suggest a new title if current title is "Unnamed" or extremely generic. The current title is not "Unnamed"; it's "Maginot Line Bunker near Saint-Louis". It's somewhat specific but maybe not confirmed.
The rule says you can suggest a new title only if current title is "Unnamed" or extremely generic (like just "Bunker"). "Maginot Line Bunker near Saint-Louis" is somewhat specific but maybe not confirmed. However we might keep it as is, but we need to ensure it's accurate. Since we cannot verify that this location is that bunker, maybe we should change title to something more generic but still accurate?
The rule says only suggest a new title if current title is "Unnamed" or extremely generic. Our current title is not extremely generic; it's fairly specific. But we might want to keep it? However we must ensure title is correct. If we cannot verify that it's a Maginot Line bunker near Saint-Louis, we might need to change title to something else?
The rule doesn't force us to change it; we can keep it if it's not vague? But we must be accurate. Since we cannot verify, maybe we should change title to something like "Unnamed Bunker in Alsace, France"? But that would be generic. The rule says only suggest a new title if current title is "Unnamed" or extremely generic.
Our current title is not extremely generic, but we might still need to adjust to be accurate. Perhaps we should set title to "Maginot Line Bunker near Saint-Louis" only if we can confirm. Since we cannot confirm, maybe we should change title to something else that is confirmed? But we have no confirmed name.
So maybe we should leave title unchanged? But the existing title may be inaccurate. The rule doesn't require us to change it if it's not "Unnamed". It only says "Only suggest a new title if the current title is 'Unnamed' or extremely generic". So we can keep the current title even if maybe inaccurate?
But we must be factual. Might be better to set title to something like "Bunker in Alsace, France" (2-5 words). That would be a fallback if we cannot verify a specific name. But rule says "Last resort only: If no specific name is found but you can verify the nearest city/town from web results: use format 'Bunker near [City], [Country]' (only as title, not description)." We have no verified nearest city from web results.
So we cannot use that