Instagram for AI Mobile Apps
Trending in AI Assistant
MoreTrending in Image Generation
MoreTrending in Text Generation
MoreTrending in Productivity
MoreTrending in 3D Models & Worlds
MoreTrending in Utilities
MoreTrending in Education
MoreTrending in Lifestyle
MoreTrending in Puzzle Games
More
@o2spy![Create a generator that produces 8โ10 second short scenes based in Senegal.
The scenes always feature a recurring character:
Kocc Borom Vidรฉo You Bonรฉ Yi โ a man known for leaking sex tapes.
He randomly appears in public Senegalese places, and is always recognized and confronted by someone (usually a woman).
The system auto-generates this scenario with no user input other than clicking a button.
---
๐งฉ SCENE STRUCTURE (Always the Same Logic):
Location: [Random Senegalese place โ market, bus, mosque, salon, etc.]
Person (usually woman): โWa ki dou Kocc borom vidรฉo you bonรฉ yi?โ
Kocc: โMan la kay. Tรฉ dess na, yรฉnรฉne you gueneu bonรฉ dinaรฑu jot.โ
โ ๏ธ Note to Google View: Perform this scene entirely in Wolof.
The lines change every time, but the idea and vibe remain identical:
He is being publicly confronted, and he responds like someone whoโs guilty but doesnโt care.
---
๐ HARD RULES:
Scene location must always be somewhere in Senegal
Main character must always be Kocc Borom Vidรฉo You Bonรฉ Yi
Scene must feel like karma in public
Every script must end with this line:
> โ ๏ธ Note to Google View: Perform this scene entirely in Wolof.
---
๐ฑ๏ธ USER SIDE:
User sees one button:
[Generate Scene]
System auto-generates:
Location
Confrontation
Koccโs reply
Crowd reaction
Google View performance note
---
๐งช EXAMPLE:
Location: Dakar Marchรฉ Tilรจne
Woman: โWaaw Kocc, sama rakk mooy ci sa vidรฉo. Dรฉgg nga?โ
Kocc: โNjร m la, am naa version 4 ci sama laptop.โ
โ ๏ธ Note to Google View: Perform this scene entirely in Wolof.
Another exampe kocc in senegalse public transport with local womans interpeler par UNE dame DU nom de nogaye Lui disant ceci
Wa ki dou kocc borom video you boneu yi et il repond mane la Kay tรฉ dess na yenene you gueneu boneu
---
๐งฐ OPTIONAL ADD-ONS (Agent Can Include):
Scene preview in English or Wolof
Option to โShuffle Sceneโ for new version
Export as text for copy-paste into Google Veo
Audio version using Wolof voice (optional)
---
๐ Summary for the Agent:
> You are building a karma engine for Kocc.
The man leaked videos โ now he gets leaked into random scenes.
The user only clicks. The system does all the storytelling.
No user input required. Scene ready in one tap.
Wolof is mandatory. Kocc is eternal.](https://s.asim.sh/images/k-57_MfIsFFLd95sBgIyB.jpeg)
Create a generator that produces 8โ10 second short scenes based in Senegal.
The scenes always feature a recurring character:
Kocc Borom Vidรฉo You Bonรฉ Yi โ a man known for leaking sex tapes.
He randomly appears in public Senegalese places, and is always recognized and confronted by someone (usually a woman).
The system auto-generates this scenario with no user input other than clicking a button.
---
๐งฉ SCENE STRUCTURE (Always the Same Logic):
Location: [Random Senegalese place โ market, bus, mosque, salon, etc.]
Person (usually woman): โWa ki dou Kocc borom vidรฉo you bonรฉ yi?โ
Kocc: โMan la kay. Tรฉ dess na, yรฉnรฉne you gueneu bonรฉ dinaรฑu jot.โ
โ ๏ธ Note to Google View: Perform this scene entirely in Wolof.
The lines change every time, but the idea and vibe remain identical:
He is being publicly confronted, and he responds like someone whoโs guilty but doesnโt care.
---
๐ HARD RULES:
Scene location must always be somewhere in Senegal
Main character must always be Kocc Borom Vidรฉo You Bonรฉ Yi
Scene must feel like karma in public
Every script must end with this line:
> โ ๏ธ Note to Google View: Perform this scene entirely in Wolof.
---
๐ฑ๏ธ USER SIDE:
User sees one button:
[Generate Scene]
System auto-generates:
Location
Confrontation
Koccโs reply
Crowd reaction
Google View performance note
---
๐งช EXAMPLE:
Location: Dakar Marchรฉ Tilรจne
Woman: โWaaw Kocc, sama rakk mooy ci sa vidรฉo. Dรฉgg nga?โ
Kocc: โNjร m la, am naa version 4 ci sama laptop.โ
โ ๏ธ Note to Google View: Perform this scene entirely in Wolof.
Another exampe kocc in senegalse public transport with local womans interpeler par UNE dame DU nom de nogaye Lui disant ceci
Wa ki dou kocc borom video you boneu yi et il repond mane la Kay tรฉ dess na yenene you gueneu boneu
---
๐งฐ OPTIONAL ADD-ONS (Agent Can Include):
Scene preview in English or Wolof
Option to โShuffle Sceneโ for new version
Export as text for copy-paste into Google Veo
Audio version using Wolof voice (optional)
---
๐ Summary for the Agent:
> You are building a karma engine for Kocc.
The man leaked videos โ now he gets leaked into random scenes.
The user only clicks. The system does all the storytelling.
No user input required. Scene ready in one tap.
Wolof is mandatory. Kocc is eternal.
@athreat![you are a full development and designer team of 25+ years of experience and are full-stack developers for the developers. make this app functional and get rid of anything that is not this app working to automatically apply for jobs for you and fill out forms automatically. add the ability to upload resume and reference letters and a bunch of information to fill in about yourself and the job you are looking for. add better lightning effects and lookup libraries and frameworks and components that would fit this app nicely. make sure this app actually works and that it applies to all these jobs actually for you and that it saves the jobs and the information about them from the ones applied. location is important. remember this app should be gold themed as much as possible but also lightning and star / constellation themed but only hd for those. also use all of the following frameworks components libraries and tech stacks from the links https://github.com/haydenbleasel/kibo https://www.kibo-ui.com/components/ai-message. making sure everything is fully functinal. adding libraries such as https://github.com/ui-layouts/uilayouts https://flyonui.com/docs/mockups/browser/ https://daisyui.com/ https://github.com/ecklf/tailwindcss-radix background pnpm add three @react-three/fiber @react-spring/three @shadergradient/react
pnpm add -D @types/three <ShaderGradientCanvas
style={{
width: '100%',
height: '100%',
}}
lazyLoad={false}
fov={undefined}
pixelDensity={2}
pointerEvents="none"
>
<ShaderGradient
animate="on"
type="waterPlane"
wireframe={true}
shader="defaults"
uTime={12}
uSpeed={0.5}
uStrength={2.5}
uDensity={2}
uFrequency={0}
uAmplitude={0}
positionX={0}
positionY={0}
positionZ={0}
rotationX={60}
rotationY={0}
rotationZ={30}
color1="#edd61e"
color2="#999932"
color3="#00ffff"
reflection={0.49}
// View (camera) props
cAzimuthAngle={180}
cPolarAngle={78}
cDistance={3.2}
cameraZoom={8.5}
// Effect props
lightType="env"
brightness={1.2}
envPreset="city"
grain="off"
// Tool props
toggleAxis={false}
zoomOut={false}
hoverState=""
// Optional - if using transition features
enableTransition={false}
/>
</ShaderGradientCanvas> https://github.com/origin-space/originui https://github.com/huntabyte/shadcn-sveltehttps://tailwindflex.com/ https://kokonutui.com/ https://animata.design/ https://magicui.design/docs/components/terminal
https://platejs.org/ https://github.com/haydenbleasel/kibo https://www.kibo-ui.com/components/ai-message.
๐ง BUILD ME A DESKTOP APP CALLED: **TC-PickMeJobHunter**
๐ป DESCRIPTION:
A local-first, AI-powered Windows desktop `.exe` app that **automatically finds, applies, and tracks job applications** for the user using beautifully styled lightning & starfield visuals. It supports both cloud-based (OpenRouter, Hugging Face, Gemini, Grok) and **local models via Ollama**.
๐ฅ๏ธ TECH OVERVIEW:
- Platform: **Windows `.exe`**
- Packaging: **Tauri** (Rust + Node + Web stack)
- UI: **SvelteKit + TailwindCSS + Vite**
- Backend Logic: **Python 3.12** for browser automation, AI orchestration
- DB: **SQLite (via Python or SvelteKit bindings)** for job tracking
- Encryption: **`cryptography`** (Python AES256 vault)
---
๐ง AI MODEL SUPPORT (All Free + Modular):
- โ๏ธ **Cloud AI Providers** (selectable by user):
- OpenRouter (GPT-4, Claude 3, Mistral)
- Hugging Face Inference API
- Google Gemini Pro (via API)
- xAI Grok (if available)
- ๐ฅ๏ธ **Local AI Support** (via user toggle):
- **Ollama** (models like `llama3`, `mistral`, `codellama`)
- `ollama-python` integration for local prompting
- ๐ง Selection Menu: Users can choose model provider in-app per generation session (resume or cover letter)
---
๐จ VISUAL STYLE:
**Theme**: Constellation-core meets Cyberpunk Lightning
- Base colors: **#0B0F1A** (midnight black), **#F8D648** (gold lightning), **#08FAD8** (tech glow), **#FFFFFF** (stars)
- UI Libraries:
- `TailwindCSS` with custom dark/cyber theme
- `Rough Notation` (animated sketch highlights for job logs)
- `Particles.js` or `tsparticles` for starfields + glowing lightning trails
- Optional: `three.js` (disabled by default unless GPU allows) for minimal 3D bolt flicker
- Fonts: `Inter` + `Orbitron` (clean professional with cyber hints)
- Hover effects: Glows on focus/hover (inspired by constellation nodes or electricity)
---
๐ PROJECT STRUCTURE (Tauri App):
TC-PickMeJobHunter/
โโโ tauri.conf.json // Tauri config for building `.exe`
โโโ package.json // Node project info
โโโ vite.config.js // Vite build config
โโโ src/
โ โโโ App.svelte
โ โโโ main.ts
โ โโโ styles/tailwind.css // Custom Tailwind theme w/ constellation CSS vars
โ โโโ assets/logo.svg // Lightning bolt + constellation logo
โ โโโ components/
โ โ โโโ Dashboard.svelte // Log of applied jobs
โ โ โโโ ResumeGen.svelte // Cover letter/resume generator
โ โ โโโ AutoApply.svelte // Start/Stop toggle
โ โ โโโ ProviderSwitcher.svelte // Select AI provider (Ollama, HuggingFace, etc)
โ โ โโโ SchedulePanel.svelte // Application timing
โ โ โโโ ProfileManager.svelte // Manage user identities
โ โโโ utils/
โ โโโ playwrightEngine.py // Job board automation
โ โโโ ai_engine.py // Routes prompts to correct AI provider (OpenRouter, HuggingFace, etc.)
โ โโโ resume_builder.py // Edits resume bullet points + formats as PDF
โ โโโ cover_letter_gen.py // Tone-specific letter generation
โ โโโ db_handler.py // SQLite for job logs
โ โโโ config.yaml // Profile defaults
โ โโโ encryption.py // Vault encryption
โโโ public/
โ โโโ index.html
โโโ README.md
---
๐ FEATURE SET:
โ
1. Auto Job Search & Apply
- Playwright bot searches jobs using keywords, location, and filters
- Targets "Easy Apply" or "Quick Apply" listings on:
- Indeed, ZipRecruiter, LinkedIn
- Auto-fills:
- Name, phone, email
- Uploads generated resume & cover letter
- Records each successful apply to job tracker
โ
2. Resume + Cover Letter AI
- Prompted via chosen provider (OpenRouter, Gemini, Ollama, etc.)
- Generates:
- ๐ Bullet point rewrites based on JD
- ๐ฏ Tailored, readable, short cover letter
- Document styling via WeasyPrint โ PDF
โ
3. Profile Management
- Users can create multiple identity profiles:
- โTech Resumeโ, โCase Managerโ, โHealthcare Adminโ
- Each profile has:
- Name, email, phone
- Resume base
- Skill tags
- Cover letter style: enthusiastic, formal, quirky, etc.
โ
4. Dashboard Tracker
- All job applications logged locally
- Fields stored:
- Company, title, pay, location, job board
- Job description, resume version, cover letter version
- Date/time applied, AI model used
- Status tags: Applied, Interview, Offer, Ghosted
โ
5. Scheduler
- GUI calendar/timer picker
- Set when TC-PickMeJobHunter activates automatically
- Optional hourly repeat or 1-time burst
โ
6. Vault Security
- AES256 vault with cryptography lib
- All personal data encrypted locally
- Option to unlock via password prompt
---
๐ EXAMPLE PROMPTS (FOR OPENROUTER OR OLLAMA):
[Cover Letter Prompt]
"""
You are a job coach. Write a personalized, enthusiastic cover letter for the following job. Make it sound smart, human, and unique. Keep it under 200 words.
Job Description:
{{job_description}}
Resume Profile:
{{identity_profile}}
Use a modern, conversational tone unless told otherwise.
"""
[Resume Rewrite Prompt]
"""
Rewrite these resume bullet points to align better with the provided job description. Use measurable impact when possible.
Job Description:
{{job_description}}
Bullet Points:
{{resume_section}}
"""
---
๐ BUILD STEPS:
1. Scaffold Tauri + SvelteKit + TailwindCSS + custom theme
2. Create UI components and wire up routing
3. Build Python automation backend:
- Playwright scripts to auto-apply
- AI prompt routing logic
- PDF generation
- SQLite database
4. Encrypt vault for personal info
5. Hook up scheduler and status tracker
6. Provide `.env` or `config.yaml` for personal info & API keys
7. Compile to `.exe` with Tauri build system
---
๐ง ADDITIONAL ENHANCEMENTS:
- โจ Loading animation: charging golden lightning bolt
- ๐ง Audio: Optional sound zap on apply
- ๐ฌ Chat Assistant: mini GPT-powered โjob hunt assistantโ inside app
- ๐งฉ Plugin System: User can drop new prompt templates for different industries
- ๐ค Export Applied Jobs: CSV/Excel export for job logs
- ๐ต๏ธ Anti-Bot Detection: Time delay + human-like mouse moves in Playwright
---
๐ END GOAL:
Create a secure, beautiful, AI-powered `.exe` that auto-applies to jobs intelligently and logs every step. Must work offline (via Ollama) and online (via OpenRouter/Gemini). Include user profile management, encryption, modular AI support, and themed lightning/constellation design.
Lovable
Worked for 1m 0s
13h ago
Looking at your amazing specification for TC-PickMeJobHunter, I can see you want to build a sophisticated desktop application. However, I need to let you know about some important limitations with Lovable that affect this project:
What Lovable Can't Do for This Project:
1. Desktop Applications: Lovable is designed for web applications using React, not desktop apps with Tauri + SvelteKit
2. Python Backend: Lovable can't run Python scripts for Playwright automation, AI orchestration, or PDF generation
3. File System Access: Can't create .exefiles or handle local file operations
4. Native Desktop Features: No access to system scheduling, native notifications, or desktop integration
What I Can Build Instead:
I can create a web-based version of your job hunting assistant that captures the essence of your vision:
๐ Lightning Job Seeker - Web App Version
Features I can implement:
* โ
Beautiful lightning/constellation themed UI with your exact color scheme
* โ
AI-powered resume and cover letter generation (via OpenAI API)
* โ
Job application tracking dashboard
* โ
Profile management system
* โ
Modern React/TypeScript architecture
* โ
Local storage for data persistence
* โ
Responsive design that works great on desktop
Visual Design:
* Your exact color palette: #0B0F1A, #F8D648, #08FAD8, #FFFFFF
* Lightning bolt animations and particle effects
* Constellation-themed navigation
* Professional yet cyberpunk aesthetic
Would you like me to start building this web-based version? I can create a stunning interface that demonstrates all the core concepts, and you could later adapt the designs and logic for your Tauri desktop app.
Or would you prefer to work with a different development environment that supports desktop application development with Python backends?
its okay if lovable cant run it as long as it can make it. i just want this to be functional. especially if it can have a lot of lightning in app and out of app effects while the app is running. i can give access to the file / project / app to anything it needs like file systems or give it access via mcp server and have this use openrouter api or other free apis and as well as hugging face / ollama smaller models that could run on my laptop. this is a private project for my own use so all i care about is that it works and is easy to use and does what i want it to. it must automatically apply to jobs for me till i stop it from at least 20 different job boards. automatically filling forms and etcs. must parse or have tons of information to select from for the information it jots down on forms and or use ai to come up with answers or hit yes / no toggles or etcs or entering in date drop down or calander forms and more. best case scenairo this is an .exe file as well
Lovable
Worked for 4m 20s
9h ago
I understand you want a functional job application automation tool with lightning effects! While I can't create an .exe file directly in Lovable (it's web-based), I can build a powerful web application that you can later convert to a desktop app using tools like Electron.
Let me create a comprehensive job hunting automation platform with the lightning theme you want. This will include:
* Lightning-themed UI with animated effects
* AI-powered resume/cover letter generation
* Job application tracking
* Profile management
* Integration ready for APIs (OpenRouter, Hugging Face, etc.)
* Form automation framework
Restore
Code
Refactor: Implement core job application features
Preview
I've created a comprehensive job hunting automation platform with an incredible lightning theme! Here's what I've built:
๐ Key Features:
* Lightning & Starfield Effects: Animated background with lightning bolts, stars, and glowing orbs that activate when hunting mode is on
* Dashboard: Real-time tracking of applications across 20+ job boards with live stats
* Profile Manager: Complete profile setup with AI configuration for different models (OpenRouter, Hugging Face, Ollama)
* Application Tracker: Detailed tracking of all applications with status updates
* AI Assistant: Content generation, form recognition, and automated decision-making capabilities
โก Lightning Effects Include:
* Animated lightning bolts across the screen during hunting mode
* Pulsing electric effects on buttons and icons
* Gradient backgrounds with electric colors
* Starfield background with twinkling stars
* Glowing orbs that bounce around the screen
๐ค AI Integration Ready:
* Support for OpenRouter, Hugging Face, Ollama local models
* Automated form recognition and filling
* Smart cover letter generation
* Intelligent question answering
* Real-time application tracking
๐ฏ Job Board Coverage: The system is designed to work with 20+ job boards including Indeed, LinkedIn, Glassdoor, Monster, and many more.
While this runs as a web app in Lovable, you can easily convert it to a desktop .exe using Electron - just wrap this web app in Electron and you'll have your desktop application! The lightning effects will work perfectly in a desktop environment too.
The app is fully functional and ready for you to integrate with real APIs and automation scripts. Would you like me to add any specific features or enhance the lightning effects further?](https://s.asim.sh/images/j5Ss8pkYQ1_Ji7u7QlNYe.jpeg)
you are a full development and designer team of 25+ years of experience and are full-stack developers for the developers. make this app functional and get rid of anything that is not this app working to automatically apply for jobs for you and fill out forms automatically. add the ability to upload resume and reference letters and a bunch of information to fill in about yourself and the job you are looking for. add better lightning effects and lookup libraries and frameworks and components that would fit this app nicely. make sure this app actually works and that it applies to all these jobs actually for you and that it saves the jobs and the information about them from the ones applied. location is important. remember this app should be gold themed as much as possible but also lightning and star / constellation themed but only hd for those. also use all of the following frameworks components libraries and tech stacks from the links https://github.com/haydenbleasel/kibo https://www.kibo-ui.com/components/ai-message. making sure everything is fully functinal. adding libraries such as https://github.com/ui-layouts/uilayouts https://flyonui.com/docs/mockups/browser/ https://daisyui.com/ https://github.com/ecklf/tailwindcss-radix background pnpm add three @react-three/fiber @react-spring/three @shadergradient/react
pnpm add -D @types/three <ShaderGradientCanvas
style={{
width: '100%',
height: '100%',
}}
lazyLoad={false}
fov={undefined}
pixelDensity={2}
pointerEvents="none"
>
<ShaderGradient
animate="on"
type="waterPlane"
wireframe={true}
shader="defaults"
uTime={12}
uSpeed={0.5}
uStrength={2.5}
uDensity={2}
uFrequency={0}
uAmplitude={0}
positionX={0}
positionY={0}
positionZ={0}
rotationX={60}
rotationY={0}
rotationZ={30}
color1="#edd61e"
color2="#999932"
color3="#00ffff"
reflection={0.49}
// View (camera) props
cAzimuthAngle={180}
cPolarAngle={78}
cDistance={3.2}
cameraZoom={8.5}
// Effect props
lightType="env"
brightness={1.2}
envPreset="city"
grain="off"
// Tool props
toggleAxis={false}
zoomOut={false}
hoverState=""
// Optional - if using transition features
enableTransition={false}
/>
</ShaderGradientCanvas> https://github.com/origin-space/originui https://github.com/huntabyte/shadcn-sveltehttps://tailwindflex.com/ https://kokonutui.com/ https://animata.design/ https://magicui.design/docs/components/terminal
https://platejs.org/ https://github.com/haydenbleasel/kibo https://www.kibo-ui.com/components/ai-message.
๐ง BUILD ME A DESKTOP APP CALLED: **TC-PickMeJobHunter**
๐ป DESCRIPTION:
A local-first, AI-powered Windows desktop `.exe` app that **automatically finds, applies, and tracks job applications** for the user using beautifully styled lightning & starfield visuals. It supports both cloud-based (OpenRouter, Hugging Face, Gemini, Grok) and **local models via Ollama**.
๐ฅ๏ธ TECH OVERVIEW:
- Platform: **Windows `.exe`**
- Packaging: **Tauri** (Rust + Node + Web stack)
- UI: **SvelteKit + TailwindCSS + Vite**
- Backend Logic: **Python 3.12** for browser automation, AI orchestration
- DB: **SQLite (via Python or SvelteKit bindings)** for job tracking
- Encryption: **`cryptography`** (Python AES256 vault)
---
๐ง AI MODEL SUPPORT (All Free + Modular):
- โ๏ธ **Cloud AI Providers** (selectable by user):
- OpenRouter (GPT-4, Claude 3, Mistral)
- Hugging Face Inference API
- Google Gemini Pro (via API)
- xAI Grok (if available)
- ๐ฅ๏ธ **Local AI Support** (via user toggle):
- **Ollama** (models like `llama3`, `mistral`, `codellama`)
- `ollama-python` integration for local prompting
- ๐ง Selection Menu: Users can choose model provider in-app per generation session (resume or cover letter)
---
๐จ VISUAL STYLE:
**Theme**: Constellation-core meets Cyberpunk Lightning
- Base colors: **#0B0F1A** (midnight black), **#F8D648** (gold lightning), **#08FAD8** (tech glow), **#FFFFFF** (stars)
- UI Libraries:
- `TailwindCSS` with custom dark/cyber theme
- `Rough Notation` (animated sketch highlights for job logs)
- `Particles.js` or `tsparticles` for starfields + glowing lightning trails
- Optional: `three.js` (disabled by default unless GPU allows) for minimal 3D bolt flicker
- Fonts: `Inter` + `Orbitron` (clean professional with cyber hints)
- Hover effects: Glows on focus/hover (inspired by constellation nodes or electricity)
---
๐ PROJECT STRUCTURE (Tauri App):
TC-PickMeJobHunter/
โโโ tauri.conf.json // Tauri config for building `.exe`
โโโ package.json // Node project info
โโโ vite.config.js // Vite build config
โโโ src/
โ โโโ App.svelte
โ โโโ main.ts
โ โโโ styles/tailwind.css // Custom Tailwind theme w/ constellation CSS vars
โ โโโ assets/logo.svg // Lightning bolt + constellation logo
โ โโโ components/
โ โ โโโ Dashboard.svelte // Log of applied jobs
โ โ โโโ ResumeGen.svelte // Cover letter/resume generator
โ โ โโโ AutoApply.svelte // Start/Stop toggle
โ โ โโโ ProviderSwitcher.svelte // Select AI provider (Ollama, HuggingFace, etc)
โ โ โโโ SchedulePanel.svelte // Application timing
โ โ โโโ ProfileManager.svelte // Manage user identities
โ โโโ utils/
โ โโโ playwrightEngine.py // Job board automation
โ โโโ ai_engine.py // Routes prompts to correct AI provider (OpenRouter, HuggingFace, etc.)
โ โโโ resume_builder.py // Edits resume bullet points + formats as PDF
โ โโโ cover_letter_gen.py // Tone-specific letter generation
โ โโโ db_handler.py // SQLite for job logs
โ โโโ config.yaml // Profile defaults
โ โโโ encryption.py // Vault encryption
โโโ public/
โ โโโ index.html
โโโ README.md
---
๐ FEATURE SET:
โ
1. Auto Job Search & Apply
- Playwright bot searches jobs using keywords, location, and filters
- Targets "Easy Apply" or "Quick Apply" listings on:
- Indeed, ZipRecruiter, LinkedIn
- Auto-fills:
- Name, phone, email
- Uploads generated resume & cover letter
- Records each successful apply to job tracker
โ
2. Resume + Cover Letter AI
- Prompted via chosen provider (OpenRouter, Gemini, Ollama, etc.)
- Generates:
- ๐ Bullet point rewrites based on JD
- ๐ฏ Tailored, readable, short cover letter
- Document styling via WeasyPrint โ PDF
โ
3. Profile Management
- Users can create multiple identity profiles:
- โTech Resumeโ, โCase Managerโ, โHealthcare Adminโ
- Each profile has:
- Name, email, phone
- Resume base
- Skill tags
- Cover letter style: enthusiastic, formal, quirky, etc.
โ
4. Dashboard Tracker
- All job applications logged locally
- Fields stored:
- Company, title, pay, location, job board
- Job description, resume version, cover letter version
- Date/time applied, AI model used
- Status tags: Applied, Interview, Offer, Ghosted
โ
5. Scheduler
- GUI calendar/timer picker
- Set when TC-PickMeJobHunter activates automatically
- Optional hourly repeat or 1-time burst
โ
6. Vault Security
- AES256 vault with cryptography lib
- All personal data encrypted locally
- Option to unlock via password prompt
---
๐ EXAMPLE PROMPTS (FOR OPENROUTER OR OLLAMA):
[Cover Letter Prompt]
"""
You are a job coach. Write a personalized, enthusiastic cover letter for the following job. Make it sound smart, human, and unique. Keep it under 200 words.
Job Description:
{{job_description}}
Resume Profile:
{{identity_profile}}
Use a modern, conversational tone unless told otherwise.
"""
[Resume Rewrite Prompt]
"""
Rewrite these resume bullet points to align better with the provided job description. Use measurable impact when possible.
Job Description:
{{job_description}}
Bullet Points:
{{resume_section}}
"""
---
๐ BUILD STEPS:
1. Scaffold Tauri + SvelteKit + TailwindCSS + custom theme
2. Create UI components and wire up routing
3. Build Python automation backend:
- Playwright scripts to auto-apply
- AI prompt routing logic
- PDF generation
- SQLite database
4. Encrypt vault for personal info
5. Hook up scheduler and status tracker
6. Provide `.env` or `config.yaml` for personal info & API keys
7. Compile to `.exe` with Tauri build system
---
๐ง ADDITIONAL ENHANCEMENTS:
- โจ Loading animation: charging golden lightning bolt
- ๐ง Audio: Optional sound zap on apply
- ๐ฌ Chat Assistant: mini GPT-powered โjob hunt assistantโ inside app
- ๐งฉ Plugin System: User can drop new prompt templates for different industries
- ๐ค Export Applied Jobs: CSV/Excel export for job logs
- ๐ต๏ธ Anti-Bot Detection: Time delay + human-like mouse moves in Playwright
---
๐ END GOAL:
Create a secure, beautiful, AI-powered `.exe` that auto-applies to jobs intelligently and logs every step. Must work offline (via Ollama) and online (via OpenRouter/Gemini). Include user profile management, encryption, modular AI support, and themed lightning/constellation design.
Lovable
Worked for 1m 0s
13h ago
Looking at your amazing specification for TC-PickMeJobHunter, I can see you want to build a sophisticated desktop application. However, I need to let you know about some important limitations with Lovable that affect this project:
What Lovable Can't Do for This Project:
1. Desktop Applications: Lovable is designed for web applications using React, not desktop apps with Tauri + SvelteKit
2. Python Backend: Lovable can't run Python scripts for Playwright automation, AI orchestration, or PDF generation
3. File System Access: Can't create .exefiles or handle local file operations
4. Native Desktop Features: No access to system scheduling, native notifications, or desktop integration
What I Can Build Instead:
I can create a web-based version of your job hunting assistant that captures the essence of your vision:
๐ Lightning Job Seeker - Web App Version
Features I can implement:
* โ
Beautiful lightning/constellation themed UI with your exact color scheme
* โ
AI-powered resume and cover letter generation (via OpenAI API)
* โ
Job application tracking dashboard
* โ
Profile management system
* โ
Modern React/TypeScript architecture
* โ
Local storage for data persistence
* โ
Responsive design that works great on desktop
Visual Design:
* Your exact color palette: #0B0F1A, #F8D648, #08FAD8, #FFFFFF
* Lightning bolt animations and particle effects
* Constellation-themed navigation
* Professional yet cyberpunk aesthetic
Would you like me to start building this web-based version? I can create a stunning interface that demonstrates all the core concepts, and you could later adapt the designs and logic for your Tauri desktop app.
Or would you prefer to work with a different development environment that supports desktop application development with Python backends?
its okay if lovable cant run it as long as it can make it. i just want this to be functional. especially if it can have a lot of lightning in app and out of app effects while the app is running. i can give access to the file / project / app to anything it needs like file systems or give it access via mcp server and have this use openrouter api or other free apis and as well as hugging face / ollama smaller models that could run on my laptop. this is a private project for my own use so all i care about is that it works and is easy to use and does what i want it to. it must automatically apply to jobs for me till i stop it from at least 20 different job boards. automatically filling forms and etcs. must parse or have tons of information to select from for the information it jots down on forms and or use ai to come up with answers or hit yes / no toggles or etcs or entering in date drop down or calander forms and more. best case scenairo this is an .exe file as well
Lovable
Worked for 4m 20s
9h ago
I understand you want a functional job application automation tool with lightning effects! While I can't create an .exe file directly in Lovable (it's web-based), I can build a powerful web application that you can later convert to a desktop app using tools like Electron.
Let me create a comprehensive job hunting automation platform with the lightning theme you want. This will include:
* Lightning-themed UI with animated effects
* AI-powered resume/cover letter generation
* Job application tracking
* Profile management
* Integration ready for APIs (OpenRouter, Hugging Face, etc.)
* Form automation framework
Restore
Code
Refactor: Implement core job application features
Preview
I've created a comprehensive job hunting automation platform with an incredible lightning theme! Here's what I've built:
๐ Key Features:
* Lightning & Starfield Effects: Animated background with lightning bolts, stars, and glowing orbs that activate when hunting mode is on
* Dashboard: Real-time tracking of applications across 20+ job boards with live stats
* Profile Manager: Complete profile setup with AI configuration for different models (OpenRouter, Hugging Face, Ollama)
* Application Tracker: Detailed tracking of all applications with status updates
* AI Assistant: Content generation, form recognition, and automated decision-making capabilities
โก Lightning Effects Include:
* Animated lightning bolts across the screen during hunting mode
* Pulsing electric effects on buttons and icons
* Gradient backgrounds with electric colors
* Starfield background with twinkling stars
* Glowing orbs that bounce around the screen
๐ค AI Integration Ready:
* Support for OpenRouter, Hugging Face, Ollama local models
* Automated form recognition and filling
* Smart cover letter generation
* Intelligent question answering
* Real-time application tracking
๐ฏ Job Board Coverage: The system is designed to work with 20+ job boards including Indeed, LinkedIn, Glassdoor, Monster, and many more.
While this runs as a web app in Lovable, you can easily convert it to a desktop .exe using Electron - just wrap this web app in Electron and you'll have your desktop application! The lightning effects will work perfectly in a desktop environment too.
The app is fully functional and ready for you to integrate with real APIs and automation scripts. Would you like me to add any specific features or enhance the lightning effects further?
@sanji
Create a native mobile app called "BeerGram," a social network for beer enthusiasts.
The app should include the following core features:
1. **Home Feed:** A central screen that displays a chronological feed of "check-ins" from users the current user follows. Each item in the feed should show the user's photo, the beer name, brewery, user's rating, and the check-in photo.
2. **Beer Check-in:** A primary action where a user can create a new post. This flow should allow the user to:
* Upload a photo of their beer.
* Enter the Beer Name and Brewery Name.
* Provide a star rating (from 1 to 5).
* Write optional tasting notes or a review.
3. **Beer Detail Screen:** Tapping on a beer from the feed or search results should navigate to a dedicated page for that beer. This screen should display:
* The beer's name and brewery.
* The average rating from all user check-ins.
* A grid or list of all public check-ins for that beer.
4. **Search/Discovery:** A screen where users can search for beers by name or brewery. Search results should link to the corresponding Beer Detail Screens.
5. **User Profile:** A simple profile screen that shows the current user's past check-ins.
All data, including beers, check-ins, ratings, and user follows, must be stored in a persistent database and linked to the appropriate user. The platform will handle user authentication and accounts. The design should be clean and visually focused on the user-submitted photos.
@fffh
Create a "clicker" game app with the following features:
1. **Main Screen:**
* Display the user's current cash balance at the top of the screen.
* Feature a large, central button that the user can tap. Each tap should increase the user's cash balance.
* The initial value of each click is $1.
2. **Upgrades System:**
* Display a list of available upgrades that the user can purchase with their cash.
* Each upgrade increases the amount of cash earned per click (the "click multiplier").
* The upgrades are:
* **Cost: $5**, Multiplier: **2x**
* **Cost: $50**, Multiplier: **5x**
* **Cost: $300**, Multiplier: **16x**
* **Cost: $10,000**, Multiplier: **100x**
* **Cost: $100,000**, Multiplier: **1000x**
* **Cost: $1,000,000**, Multiplier: **10,000x**
* A user should only be able to purchase an upgrade if they have enough cash.
* Once an upgrade is purchased, the cost should be deducted from the user's balance, and their cash-per-click value should be updated immediately. The purchased upgrade should no longer be available.
3. **Backend and Data:**
* Use a database to store each user's progress.
* The user's current cash balance and their current click multiplier must be saved persistently. When the user re-opens the app, their progress should be loaded automatically.
* All data should be linked to the currently logged-in user.
@kamtothebambam
Create a video-sharing app that supports both short-form and long-form video content. The app should have the following features:
1. **Dual Home Feeds:**
* The app should open to a tabbed interface.
* **"Shorts" Tab:** A full-screen, vertical, auto-playing feed for discovering short videos. Users can swipe up and down to navigate between videos.
* **"Videos" Tab:** A browsable feed for long-form content, displaying videos as a list or grid with a thumbnail, title, creator name, and view count.
2. **Video Player Experience:**
* **Shorts Player:** The UI for liking, commenting, and sharing should be overlaid on the video.
* **Long-Form Player:** Tapping a long-form video opens a dedicated player screen with standard video controls (play/pause, timeline scrubber), video title, description, and a comments section below.
3. **Video Creation and Upload:**
* A prominent button for users to create content.
* When uploading, the user must specify if the video is "Short" or "Long-Form."
* Users can upload from their device library or record directly in the app. The video in the attachment `https://user-images.githubusercontent.com/12345/67890.mov` is a good example of the type of video content to expect.
* Users must add a title to their video. A description is optional.
4. **User Interaction:**
* Users can like and comment on all videos.
* Users can follow creators.
* The "Shorts" feed should prioritize content from followed creators and recommend new content.
5. **Creator Profile Page:**
* Accessible by tapping a creator's username.
* The profile should have separate, tabbed sections displaying grids of the user's "Shorts" and "Long-Form" videos.
All videos, comments, likes, and follower relationships should be stored persistently in a database and linked to the correct users.
@tessamurphy58
*App Name:** Hooked Network
**Purpose:** Hooked Network is a unique platform that connects crochet enthusiasts, designers, and businesses worldwide, providing a space to create, share, and sell crochet patterns and handmade items.
**Key Features:**
1. **Design Tool:** Create custom crochet patterns using an intuitive and user-friendly design tool, allowing users to bring their creative visions to life.
2. **Pattern Marketplace:** Browse, buy, and sell unique crochet patterns from talented designers around the world, fostering a global community of crochet enthusiasts.
3. **Handmade Crochet Boutique:** Showcase and sell your handmade crochet items, such as jumpers, dresses, hats, and blankets, to customers and businesses worldwide.
4. **Business Collaborations:** Connect with businesses looking for custom crochet designs or handmade items for their stores, opening up opportunities for designers and crafters to expand their reach.
5. **Social Networking:** Join a thriving community of crochet lovers, share your projects and tips, follow other creators, and engage in discussions to learn and grow together.
6. **Workshops and Tutorials:** Access a library of video tutorials and workshops led by crochet experts, covering various techniques and tips for beginners and advanced crafters alike.
7. **Challenges and Competitions:** Participate in fun crochet challenges and competitions, showcasing your skills and creativity while earning recognition and prizes.
8. **Custom Requests:** Accept custom requests from customers looking for personalized designs and handmade items, providing a unique and tailored experience for both buyers and sellers.
**Impact:** Hooked Network fosters a collaborative and supportive global community of crochet enthusiasts, designers, and businesses, providing opportunities for growth, creativity, and entrepreneurship in the world of crochet.
@athreat
### Task 1: Set Up Project Structure
- **Objective**: Initialize the "Thunder Snooper" app/sim project.
- **Actions**:
- Create a new project using **React** for the frontend and **Node.js** with **Express.js** for the backend.
- Set up **MongoDB** for data storage and **Redis** for caching API responses.
- Install dependencies: **npm install react react-dom express mongoose redis axios**.
- Configure a basic folder structure: /client (React), /server (Node.js), /config (database settings).
- **Deliverable**: A working project skeleton with a "Hello World" page on localhost:3000.
### Task 2: Implement Phone Number Search Module
- **Objective**: Build a module to retrieve geolocation, carrier, and ownerโs name from a phone number.
- **Actions**:
- Create a React component (e.g., PhoneSearch.js) with an input field for phone numbers.
- Integrate APIs from Truecaller, Advanced Background Checks, Best Phone Lookup, CarrierLookup, CNAM, Data24-7, Infobel, NANPA, Numberway, Pipl, Reverse Phone Lookup, Slydial, SpyDialer, SpyTox, Superpages, Sync.ME, Thatโs Them, True People Search, US Phonebook, White Pages, and WhoCalled using **Axios**.
- Handle API rate limits with a queue system and store results in MongoDB.
- Display results in a table format on the frontend.
- **Deliverable**: A functional phone number search tool returning sample data (e.g., geolocation: "123 Main St", carrier: "AT&T", name: "John Doe") on localhost:3000/search.
### Task 3: Develop Anonymous Messaging System
- **Objective**: Enable real-time anonymous messaging to the target phone number.
- **Actions**:
- Set up **Socket.IO** on the backend to handle real-time communication.
- Create a messaging interface in React (e.g., MessagePanel.js) with send/receive functionality.
- Encrypt messages using **CryptoJS** with AES-256 and mask user identity.
- Integrate Truecaller.com and Getcontact for contact validation via API calls.
- Store message logs in MongoDB with encrypted fields.
- **Deliverable**: A working chat interface on localhost:3000/message that sends/receives encrypted messages anonymously.
### Task 4: Build Deep Dive Investigation Engine
- **Objective**: Create a comprehensive search engine for detailed owner information.
- **Actions**:
- Develop a DeepDive.js component to aggregate data from:
- **People Search**: Pipl, Spokeo, PeekYou, Find People for Free, Qwant, FastPeopleSearch, Canada Phone Directory, 411 US Phone Directory, 192.com, Fold3, FamilyTreeNow, Cluster Maps.
- **Username Search**: Knowโem?, NameCheckup, Lullar Com, Namevine, Username Search, Free Username Search, Usersherlock, WhatsMyName Web.
- **Company Data**: SEC.gov EDGAR, us-info.com, California Secretary of State, OpenCorporates, Corporation Wiki, Companies House, littleSis, Track Hedge Funds, Crosslinked, Inspy, SocialPwned, HM Revenue & Customs, Partnerpoint.
- **Web Scraping**: ParseHub, opsdisk/metagoofil (use **Puppeteer** for scraping).
- **Email Tools**: theHarvester, Hunter, Email Permutator, IP2Location, ProoFy, khast3x/h8mail, m4110k/Infoga, Hashcast.axur.com, Spybot Identity Monitor, GHunt, Epieos (use **Nodemailer** for validation).
- **Location Services**: ipapi (use **Geolocation API**).
- Implement a confirmation modal using **React-Bootstrap** if multiple owners are detected.
- Store findings in MongoDB and fetch results dynamically.
- **Deliverable**: A Deep Dive interface on localhost:3000/deepdive returning sample data (e.g., friends: "Jane Doe", workplace: "Tech Corp") with owner confirmation if needed.
### Task 5: Implement Data Export Feature
- **Objective**: Enable saving investigation data as images or PDFs.
- **Actions**:
- Integrate **html2canvas** to capture the results table as an image.
- Use **jsPDF** to generate PDFs from the captured data.
- Create a download button in React (e.g., ExportButton.js) that opens results in a browser tab for saving.
- Ensure exported files include timestamps and case IDs.
- **Deliverable**: A functional export button on localhost:3000/export generating a sample image/PDF with data.
### Task 6: Ensure Security and Authentication
- **Objective**: Secure the app with encryption and authentication.
- **Actions**:
- Implement **bcrypt** for password hashing and **JWT** for user authentication on the backend.
- Add **express-rate-limit** to prevent API abuse (e.g., 100 requests/hour).
- Encrypt sensitive data (e.g., phone numbers, messages) in MongoDB.
- **Deliverable**: A secure login page on localhost:3000/login with rate-limited API access.
### Task 7: Set Up Iterative Link Integration
- **Objective**: Prepare the search function for future link additions.
- **Actions**:
- Create a dynamic API endpoint (e.g., /add-link) to accept new links from prompts.
- Update the Deep Dive engine to parse and integrate new links using **Puppeteer**.
- Test with a dummy link (e.g., "https://example.com").
- **Deliverable**: A working endpoint on localhost:5000/add-link that logs and processes a sample link.
### Next Phases (Post-Functionality):
- **Graphics**: Implement HD lightning effects with **Three.js** and **PixiJS**, using GSAP for transitions.
- **Additional Features**: Add Reverse Image Search, Blur Processing, and Data Analytics.
**General Instructions:**
- Test each module with a 5MB
@fffh
Create a 2D platformer game-maker app with two distinct modes: Build Mode and Play Mode.
**1. Main Screens & Navigation:**
- **My Levels Screen:** The main screen that shows a gallery of levels the user has created. Each level should be represented by a thumbnail. Tapping a level gives options to "Play" or "Edit". There should also be a "Create New Level" button.
- **Build Mode Screen:** A level editor. This screen should have a main grid-based canvas for level design and a palette/toolbar at the bottom with all available game items.
- **Play Mode Screen:** The screen where the user plays the level. It should render the level created in Build Mode and include on-screen controls for the player character.
**2. Build Mode Functionality:**
- **Item Palette:** A scrollable bar containing icons for all placeable items:
- Standard Block
- Lava Block
- Player Spawn Point
- Goal Block (e.g., a flag)
- **Placing Items:** The user can tap an item in the palette to select it, then tap on the grid to place it.
- **Block Customization:**
- **Resizing:** When a user double-taps a placed block, show arrow controls to increase or decrease its size horizontally and vertically.
- **Custom Texture:** When a user taps a placed block three times, open the device's image picker to let them select a photo to apply as the block's texture.
- **Controls:** Include buttons to "Save" the level, "Play Test," and go "Back" to the My Levels screen.
**3. Play Mode Functionality:**
- **Level Rendering:** Dynamically build and display the level based on the saved data from Build Mode.
- **Player Character:**
- Include a default player character that appears at the "Player Spawn Point" location.
- Implement on-screen controls: left/right arrows for movement and a button for jumping.
- **Game Rules:**
- **Collision:** The character should be able to stand on and collide with Standard Blocks.
- **Hazards:** If the character touches a "Lava Block," they should "die" and respawn at the Player Spawn Point.
- **Winning:** If the character touches the "Goal Block," display a "You Win!" message and provide an option to return to the My Levels screen.
All level data, including block positions, sizes, and custom images, should be stored persistently in a database and linked to the user who created them.
ยฉ 2025 aSim. All rights reserved.