We need to output JSON with fields: joke_text, headline, meme_caption, image_pro...

We need to output JSON with fields: joke_text, headline, meme_caption, image_prompt, meta_title, meta_description. Must be German language, no markdown code blocks. Must be unique, start with Chuck Norris or mention him. Provide a classic Chuck Norris joke. Must be funny, unique, typical. Must avoid repeats. Provide image prompt in English describing scene. Provide headline max 80 chars, meme_caption max 80 chars, meta_title max 60 chars, meta_description max 160 chars. Ensure JSON format, no extra text. No markdown. Provide fields exactly as requested. We need to craft a joke: maybe "Chuck Norris can divide by zero." But that's common. Need unique. Maybe "Chuck Norris can make a round square." Or "Chuck Norris can hear sign language." Hmm. We need to start with "Chuck Norris" or mention him. Could be "Chuck Norris can make a horse go faster by staring at it." Not sure. Better: "Chuck Norris can solve a Rubik's Cube blindfolded while skydiving." That's typical but maybe okay. But must be unique, not repeat existing jokes. Let's think of a fresh one: "Chuck Norris can make a snowflake melt just by looking at it." That's unique. But need to be classic Chuck Norris style: exaggeration of superhuman abilities. Maybe: "Chuck Norris can make a computer understand sarcasm just by blinking." That's okay. But we need to ensure uniqueness. Let's craft: "Chuck Norris can make a traffic light turn green just by winking." That's classic. But we need to ensure it's not a known joke. I think it's okay. Now produce JSON. Fields: joke_text: full joke text in German, starting with "Chuck Norris". Should be a full sentence maybe. headline: short headline max 80 chars. meme_caption: short caption max 80 chars. image_prompt: English prompt describing specific scene from the joke, hyper-realistic portrait etc. Must include [Spezifische Szene aus dem Witz] placeholder? The instruction says: "A hyper-realistic portrait of a tough, legendary action hero man in his 50s with reddish-brown hair, thick reddish beard, piercing eyes, muscular build. Black sleeveless vest. [Spezifische Szene aus dem Witz]. Photorealistic action movie poster, dramatic cinematic lighting, American flag background. No text, no captions." So we need to embed the specific scene from the joke into the prompt. For example, "winking at a traffic light causing it to turn green". So we can write: "winking at a traffic light causing it to turn green". So full prompt: "A hyper-realistic portrait of a tough, legendary action hero man in his 50s with reddish-brown hair, thick reddish beard, piercing eyes, muscular build. Black sleeveless vest. winking at a traffic light causing it to turn green. Photorealistic action movie poster, dramatic cinematic lighting, American flag background. No text, no captions." Make sure it's English. meta_title: SEO Title max 60 chars. Something like "Chuck Norris lässt Ampel grün werden – Witz". Count characters: "Chuck Norris lässt Ampel grün werden – Witz" length? Let's count: Chuck(5) + space=1 =>6, Norris(6)=>12, space=13, lässt(5)=>18, space=19, Ampel(5)=>24, space=25, grün(4)=>29, space=30, werden(6)=>36, space=37, – (1?) maybe dash counts, space=38, Witz(4)=>42. So under 60. Good. meta_description: max 160 chars. Something like "Ein klassischer Chuck Norris Witz, bei dem er mit einem Zwinkern eine Ampel zum Grünschalten bringt. Lachen garantiert!" Count characters: let's approximate <160. Now ensure JSON fields are correct. joke_text: Must be full joke text. Could be: "Chuck Norris kann mit einem einzigen Zwinkern eine Ampel auf Grün stellen, selbst wenn sie rot ist." That's German, starts with Chuck Norris. That's okay. But need to be unique. Let's craft: "Chuck Norris kann mit einem einzigen Zwinkern eine Ampel auf Grün stellen, selbst wenn sie rot ist." That's a joke. Now headline: maybe "Chuck Norris lässt Ampel grün werden". That's under 80 chars. meme_caption: maybe "Wenn Chuck Norris winkt, wird die Ampel grün!" That's under 80. Now produce JSON. Check lengths: headline length: "Chuck Norris lässt Ampel grün werden" count: Chuck(5)+space=6+Norris(6)=12+space=13+lässt(5)=18+space=19+Ampel(5)=24+space=25+grün(4)=29+space=30+werden(6)=36. So 36 chars, okay. meme_caption: "Wenn Chuck Norris winkt, wird die Ampel grün!" Count: Wenn(4)+space=5+Chuck(5)=10+space=11+Norris(6)=17+space=18+winkt,(6 inc comma?) winkt, is 6 inc comma? Actually "winkt," 6 characters inc comma? w i n k t , =6. So 18+6=24+space=25+die(3)=28+space=29+Ampel(5)=34+space=35+grün!(5 inc exclamation) =40. So under 80. Now meta_title: "Chuck Norris lässt Ampel grün werden – Witz" length approx 42, okay. meta_description: "Ein klassischer Chuck Norris Witz, bei dem er mit einem Zwinkern eine Ampel zum Grünschalten bringt. Lachen garantiert!" Let's count roughly: "Ein klassischer Chuck Norris Witz, bei dem er mit einem Zwinkern eine Ampel zum Grünschalten bringt. Lachen garantiert!" That's likely under 160. Let's count quickly: We'll approximate it's okay. Now image_prompt: Must be English, include scene: "winking at a traffic light causing it to turn green". Provide full prompt as described. Now ensure JSON format: no extra text, just JSON object. Let's craft final JSON: { "joke_text": "Chuck Norris kann mit einem einzigen Zwinkern eine Ampel auf Grün stellen, selbst wenn sie rot ist.", "headline": "Chuck Norris lässt Ampel grün werden", "meme_caption": "Wenn Chuck Norris winkt, wird die Amp