From Wikiprompt, the free prompt encyclopedia
Wikiprompt is a collaborative encyclopedia of AI prompts, featuring curated templates for various AI tools including ChatGPT, Midjourney, DALL-E, Claude, and more. It contains 3213 prompts by 225 community members.
All, Creative, Marketing, Business, Productivity, Coding, Education, Personal, Research, Other
All, GPT Image 2 (800), ChatGPT (298), Model-agnostic (247), Seedance 2.0 (242), Nano Banana Pro (226), Gemini Nano Banana (210), Gemini Nano Banana Pro (182), Grok Imagine (83), Claude (61), Gemini (44), Midjourney v7 (29), Nano Banana 2 (27), Veo 3 (25), Gemini Nano Banana 2 (24), GPT Image 1.5 (24), Midjourney (21), NotebookLM (20), Midjourney v8.1 (14), ChatGPT / Gemini (12), Kling AI 3.0 Turbo (12)
All, cinematic (198), photorealistic (115), editorial (106), portrait (96), commercial (39), fashion (38), minimalist (31), studio (31), luxury (30), product-photography (29), lifestyle (28), golden-hour (26), surreal (25), miniature (23), infographic (21), 3d-render (20), poster (20), candid (19), monochrome (19), macro (18)
All, Creative generation (2567), Template (321), One-shot (226), System prompt (50), Workflow (39), Persona (6), Agent (3)
624 prompts
One-shot prompt for a cinematic 3D superhero battle scene with Avengers style characters, built with Claude.

One-shot prompt recreating the Star Wars Duel of the Fates lightsaber battle in 3D. Posted as a same-prompt model comparison.

A one-shot ChatGPT image prompt that mashes up a political figure with a classic movie poster parody.
One-shot prompt animating an artist sketching a hyper-realistic Mona Lisa portrait, built with Claude.

One-shot prompt for a detailed 3D police chase scene with multiple cruisers, sirens, and a helicopter, built with Claude.

A luxury Arabic-language template prompt for designing a professional medical rubber stamp for an ENT clinic, with clean typography and a realistic paper-stamp texture.

One-shot prompt animating a log cabin being built in real time in the Wild West, made with Claude.

A fun scribbly-art Midjourney style-reference blend template for quick creative experiments and art direction, using a fixed --sref code stack.

A short text-to-image prompt (originally in Arabic) for a realistic Arabic calligraphy artwork centered on premium paper with an ink pen and warm cinematic lighting; swap in any Arabic phrase.

A short cinematic Seedance 2.0 video prompt where a fired arrow trails endless fabric and triggers empty land to morph into a modern housing development.

A short Pixar-style 3D character beat: a girl struggles with her hair, then triumphs before heading out. Made with Seedance 2.0.

A short one-line prompt that produces a minimalist gouache-style illustration with matte flat color blocks and a muted retro palette on cream. Swap [SUBJECT].

A short video prompt where a camera glides through an empty apartment as glowing golden wireframes materialize into real furniture. Made with Seedance 2.0.

One-shot prompt for a detailed medieval castle siege with a battering ram, siege tower, and archers. Posted as a which-model-did-it-better comparison.

One-shot prompt for a 3D animation of horseback robbers holding up a train in the Wild West, built with Claude.

A vision to code prompt that has Claude Fable 5 rebuild a working front-end or data table directly from screenshots.
A verification prompt that has Claude Fable 5 spawn an independent fresh-context subagent to check its own work against a spec instead of self-grading.
A guardrail template that scopes an autonomous coding agent's permissions and forces honest status reporting, reducing reward hacking behavior.
A command code you paste once at the start of a chat so that any message ending in /ATTACK gets argued against as hard as the model honestly can, instead of getting agreement.
A command code that makes the model predict the most likely way a plan goes wrong before you commit to it, whenever a message ends in /FAILHOW.