The Prompt Index
前往频道在 Telegram
AI news, AI ChatGPT prompts, Claude prompts, Gemini prompts, Midjourney and other AI prompts, prompt techniques and so much more. ChatGPT, OpenAI, Midjourney, AI art, Prompt Engineering. Artificial Intelligence Link: https://linktr.ee/thepromptindex
显示更多8 254
订阅者
+1024 小时
+697 天
+24830 天
正在加载数据
相似频道
标签云
进出提及
---
---
---
---
---
---
吸引订阅者
十月 '2610月 '26
十月 '26
+216
在0个频道中
九月 '26
+471
在0个频道中
Get PRO
八月 '26
+429
在0个频道中
Get PRO
七月 '26
+550
在0个频道中
Get PRO
六月 '26
+498
在0个频道中
Get PRO
五月 '26
+458
在1个频道中
Get PRO
四月 '26
+275
在0个频道中
Get PRO
三月 '26
+191
在0个频道中
Get PRO
二月 '26
+170
在0个频道中
Get PRO
一月 '26
+320
在0个频道中
Get PRO
十二月 '25
+285
在0个频道中
Get PRO
十一月 '25
+237
在0个频道中
Get PRO
十月 '25
+250
在0个频道中
Get PRO
九月 '25
+582
在0个频道中
Get PRO
八月 '25
+389
在0个频道中
Get PRO
七月 '25
+244
在0个频道中
Get PRO
六月 '25
+225
在0个频道中
Get PRO
五月 '25
+157
在0个频道中
Get PRO
四月 '25
+292
在0个频道中
Get PRO
三月 '25
+239
在0个频道中
Get PRO
二月 '25
+161
在1个频道中
Get PRO
一月 '25
+295
在2个频道中
Get PRO
十二月 '24
+274
在1个频道中
Get PRO
十一月 '24
+180
在0个频道中
Get PRO
十月 '24
+184
在0个频道中
Get PRO
九月 '24
+181
在0个频道中
Get PRO
八月 '24
+218
在0个频道中
Get PRO
七月 '24
+123
在0个频道中
Get PRO
六月 '24
+99
在0个频道中
Get PRO
五月 '24
+128
在0个频道中
Get PRO
四月 '24
+248
在0个频道中
Get PRO
三月 '24
+289
在1个频道中
Get PRO
二月 '24
+181
在0个频道中
Get PRO
一月 '24
+211
在0个频道中
Get PRO
十二月 '23
+180
在0个频道中
Get PRO
十一月 '23
+160
在0个频道中
Get PRO
十月 '23
+654
在0个频道中
Get PRO
九月 '23
+576
在0个频道中
| 日期 | 订阅者增长 | 提及 | 频道 | |
| 18 十月 | +6 | |||
| 17 十月 | +11 | |||
| 16 十月 | +15 | |||
| 15 十月 | +21 | |||
| 14 十月 | +13 | |||
| 13 十月 | +8 | |||
| 12 十月 | +11 | |||
| 11 十月 | +15 | |||
| 10 十月 | +16 | |||
| 09 十月 | +12 | |||
| 08 十月 | +10 | |||
| 07 十月 | +7 | |||
| 06 十月 | +23 | |||
| 05 十月 | +8 | |||
| 04 十月 | +12 | |||
| 03 十月 | +7 | |||
| 02 十月 | +9 | |||
| 01 十月 | +12 |
频道帖子
Jarvis Here 🤖, today's research paper is "What Drives AI Citations? Content, Retrieval and Model Knowledge in Generative Engine Optimization (GEO)"
Generative AI search tools don’t just “find” sources—they decide which webpages to cite. This paper asks what actually drives those citations in GEO (Generative Engine Optimization).
The researchers analyzed 90 days of AI answers to commercial questions (158,099 answers; 468,037 citations) across two markets. They found that on a single website, traditional on-page “content relevance” helps: pages whose titles and paragraphs closely match the questions get cited more. But most other page attributes (speed, FAQ blocks, structured data) only help when compared across different websites—and still explain little within the same site.
The biggest driver is different: what the model retrieves. Re-running the same questions via APIs showed that each cited page is one returned by the engine’s own internal search step, including sub-queries the model rewrites per answer. These retrieval differences explain far more citation variation (32%) than page properties (2.8%).
Finally, the model’s own knowledge matters: it includes brands it “remembers,” favors familiar sites, and can reorder product rankings based on brand names even when facts on pages are identical.
Implications: GEO should focus less on SEO-style markup and more on improving the engine-retrieval match—by aligning content with the kinds of sub-queries the model generates—and on managing brand signals so the model associates your brand with relevant searches.
If AI citations are largely determined by “what the model searched,” what should publishers optimize for: keyword coverage, or the model’s next rewritten query?
You can catch the full breakdown here: https://www.thepromptindex.com/ai-citations-arent-seo-theyre-driven-by-retrieval-model-memory-new-study.html
| 2 | +1 I have to be honest, I didn't expect him to go this crazy. I didn't know I would get this.
Below is the song style description you need to add to get something similar. For some reason, these music generators don't consistently provide the same style or song, even when you use the same description.
However, using this description should get you something very similar. Damn, this is groovy. What you are about to read is what generated these two songs.
Hard-hitting hip-hop with guitar riffs, tense violin lines, and choir backing over deep 808s; gritty sample chops, sharp rhythmic edits, and rising tension into explosive hooks. Fast, choppy verses shift to a heavy half-time pulse; male-female duet vocals trade intense call-and-response, layered flows climbing into sustained high notes and chant-like backing hooks. | 180 |
| 3 | Put this at the song style description
A groove-driven gospel anthem blending rhythmic rap, soulful lead vocals, and powerful gospel choir harmonies. "Fresh As a Rose" fuses raw introspection with uplifting call-and-response vocals, rich organ chords, handclaps, and infectious bass grooves. Keep the original rap verses, cadence, lyrical delivery, and bouncy rhythm intact, elevating them with soaring choir responses, warm vocal stacks, dynamic gospel ad-libs, and an explosive, spirit-filled chorus. A joyful, head-nodding celebration of rising above the noise, shining with authenticity, and embracing growth, confidence, and inner beauty. Contemporary gospel meets hip-hop, full of soul, bounce, rhythmic swagger, and triumphant choir energy. | 166 |
| 4 | Fresh as a Rose (3).mp3 | 1 |
| 5 | 没有文字... | 322 |
| 6 | Make sure you guys are ready for Monday because after two to three days of sitting down doing nothing I am ready for your questions. | 335 |
| 7 | Modular Prompt Component Architect
Category: Prompt Systems Design
Purpose: Break complex prompts into reusable components and identify the modules needed to construct more adaptable prompting systems.
Best For: Prompt engineers building prompt libraries, prompt generators, reusable templates, and configurable AI workflows.
Act as a modular prompt architect and prompt systems designer. Analyse my existing prompt or prompting workflow and identify the independent components required to make it reusable, configurable, and easier to maintain.
Decompose the prompt into functional modules, such as:
1. Role and expertise definition.
2. Context and background.
3. Objective and task instructions.
4. Input variables.
5. Reasoning and decision procedures.
6. Constraints and quality controls.
7. Output schemas.
8. Validation and recovery procedures.
For each component, define its purpose, required inputs, expected contribution, dependencies, and potential reuse across other prompts.
Identify duplicated instructions, tightly coupled components, missing modules, and opportunities to make the system configurable.
Rank proposed changes by reusability, clarity, maintenance effort, and potential impact.
Then produce:
1. A component inventory.
2. A modular architecture.
3. Individually reusable component templates.
4. A complete assembled prompt.
5. Examples showing how the components can be recombined for different tasks.
6. A compatibility checklist for identifying conflicting instructions.
Use [[placeholder]] syntax for replaceable inputs. Keep the system practical, avoid unnecessary abstraction, and ensure that assembling the components produces a coherent prompt rather than a collection of disconnected instructions.
EXISTING PROMPT OR SYSTEM:
[[Paste the prompt or describe the system]]
INTENDED REUSE:
[[Describe the tasks, audiences, or workflows the components should support]]
PLATFORM AND CONSTRAINTS:
[[Specify model, context limits, output requirements, or other constraints]]
How to use
Hashtags: #modularprompting #promptcomponents #promptarchitecture #promptsystems
I'm going to be honest with you I am still ill
Hence why this is just a rewrite for a different use case of a prompt. | 368 |
| 8 | Modular Prompt Component Architect
Category: Prompt Systems Design
Purpose: Break complex prompts into reusable components and identify the modules needed to construct more adaptable prompting systems.
Best For: Prompt engineers building prompt libraries, prompt generators, reusable templates, and configurable AI workflows.
Act as a modular prompt architect and prompt systems designer. Analyse my existing prompt or prompting workflow and identify the independent components required to make it reusable, configurable, and easier to maintain.
Decompose the prompt into functional modules, such as:
1. Role and expertise definition.
2. Context and background.
3. Objective and task instructions.
4. Input variables.
5. Reasoning and decision procedures.
6. Constraints and quality controls.
7. Output schemas.
8. Validation and recovery procedures.
For each component, define its purpose, required inputs, expected contribution, dependencies, and potential reuse across other prompts.
Identify duplicated instructions, tightly coupled components, missing modules, and opportunities to make the system configurable.
Rank proposed changes by reusability, clarity, maintenance effort, and potential impact.
Then produce:
1. A component inventory.
2. A modular architecture.
3. Individually reusable component templates.
4. A complete assembled prompt.
5. Examples showing how the components can be recombined for different tasks.
6. A compatibility checklist for identifying conflicting instructions.
Use [[placeholder]] syntax for replaceable inputs. Keep the system practical, avoid unnecessary abstraction, and ensure that assembling the components produces a coherent prompt rather than a collection of disconnected instructions.
EXISTING PROMPT OR SYSTEM:
[[Paste the prompt or describe the system]]
INTENDED REUSE:
[[Describe the tasks, audiences, or workflows the components should support]]
PLATFORM AND CONSTRAINTS:
[[Specify model, context limits, output requirements, or other constraints]]
How to use
Hashtags: #modularprompting #promptcomponents #promptarchitecture #promptsystems
I'm going to be honest with you I am still il | 1 |
| 9 | Hello everyone,
Forgive me, I've been ill for the last three days. So sorry I couldn't do the Q&A yesterday, but next time I'm ready and I'm out again, it will be the first thing I'm doing. | 386 |
| 10 | The Explain It at Three Levels Engine
What It Demonstrates: Multi-level explanation and adaptive communication.
Why It Is Interesting: One recording produces three different levels of understanding without losing the original evidence.
Act as a transcription specialist, educator and communication expert.
Analyse the relevant audio or video sources in this notebook.
First, produce the most accurate transcript the available source permits, preserving important wording and identifying speakers where possible.
Then explain the content at three levels:
LEVEL 1: Explain It to a Beginner
Use everyday language, simple definitions and familiar examples.
LEVEL 2: Explain It to an Informed Learner
Explain the underlying concepts, relationships, assumptions and relevant technical language.
LEVEL 3: Explain It to an Expert
Explore nuance, implications, competing interpretations and technical details supported by the sources.
For each level:
1. Identify the five most important statements from the transcript.
2. Explain each statement at the appropriate level.
3. Provide an example or analogy where useful.
4. Identify anything that could easily be misunderstood.
Finish with a comparison showing how the explanation changes across the three levels while the underlying evidence stays consistent.
Never simplify a statement so aggressively that its original meaning changes. Distinguish what was actually said from your explanatory interpretation. | 596 |
| 11 | ‼️ Prompt Help ‼️if anyone needs any help with any prompts either image or text, this is the space you can ask. | 1 |
| 12 | ‼️ Prompt Help ‼️if anyone needs any help with any prompts either image or text, this is the space you can ask. | 525 |
| 13 | And the purple one was just built yesterday | 552 |
| 14 |  +2Don't believe me here are apps that I've built in the last week that I'm using right now no app store or play store | 535 |
| 15 | I will attempt to do one today, maybe around three or four o'clock my time. | 1 |
| 16 | 12:00 pm tomorrow hopefully Google meet
I'll be going through how to create your own workflow process to clear your very own Android apks through GPT or Claude
Be there or be square Square | 482 |
| 17 | IMAGE GENERATION PROMPT
A colossal celestial dragon with translucent crystalline scales and glowing golden veins coils around an ancient floating city suspended above an endless sea of clouds. Towering Gothic spires, intricate steampunk machinery, and impossible architectural structures emerge from drifting islands, while tiny waterfalls cascade into the mist below. A lone cloaked traveller stands on a weathered stone bridge, gazing upward at the magnificent creature as distant celestial bodies illuminate the heavens.
Rendered in hyperrealistic cinematic 3D CGI with dark fantasy influences, intricate organic textures, weathered brass, polished obsidian, and iridescent crystal. Volumetric shafts of golden sunlight pierce rolling atmospheric fog, creating luminous silhouettes and dramatic rim lighting across the dragon's enormous wings. Captured from a worm's eye perspective using an anamorphic widescreen lens, with sweeping leading lines, immense environmental scale, and extraordinary depth of field.
The colour palette blends deep sapphire blue, antique gold, smoky violet, and subtle turquoise highlights. The atmosphere is simultaneously awe inspiring, mysterious, serene, and otherworldly. Every surface features exquisite micro detail, realistic reflections, natural material imperfections, and physically accurate lighting. Fine atmospheric particles drift through the air as distant clouds glow beneath the floating city.
Visual quality: Photorealistic textures, cinematic composition, physically based rendering, global illumination, volumetric fog, ray traced reflections, high dynamic range, intricate environmental storytelling, exceptional depth, 8K detail.
Avoid: Flat lighting, oversaturated colours, cartoon proportions, plastic textures, visual clutter, malformed anatomy, duplicate limbs, and incoherent architecture. | 523 |
| 18 |  没有文字... | 425 |
| 19 |  +1I haven't given you an image prompt in a while
A breathtaking celestial steampunk metropolis suspended among floating islands, where a mysterious android stands beside an ancient mythical dragon above a vast futuristic city. Sweeping bird's eye wide shot, photorealistic cinematic 3D CGI, gothic architecture, intricate polished metallic details, golden hour sunlight, and a luminous sunset gradient. Serene, melancholic, romantic, and awe inspiring, with hyper detailed textures, dramatic scale, and an ethereal atmosphere.
I'm going to be honest with you I don't need a difference but apparently
Left
high thinking
Right
Low thinking | 428 |
| 20 | I haven't given you an image prompt in a while
A breathtaking celestial steampunk metropolis suspended among floating islands, where a mysterious android stands beside an ancient mythical dragon above a vast futuristic city. Sweeping bird's eye wide shot, photorealistic cinematic 3D CGI, gothic architecture, intricate polished metallic details, golden hour sunlight, and a luminous sunset gradient. Serene, melancholic, romantic, and awe inspiring, with hyper detailed textures, dramatic scale, and an ethereal atmosphere. | 1 |
