I Built a Photoshop AI Agent in n8n with no code (NanoBanana)

I Built a Photoshop AI Agent in n8n with no code (NanoBanana)

🎙 Nate Herk 👥 964K 📅 September 5, 2025 ⏱ 17 min 👁 88K 📄 tutorial 🧭 2026-08-28
Available in: English (current) Français

Keywords

n8nAI agentNano Bananaimage editingno-codeGoogle DriveTelegramFAL AI

Summary

In this tutorial, Nate Herk demonstrates how to build a no-code Photoshop AI agent using n8n, powered by Google’s Nano Banana image generation model. The agent can combine images, edit them, and manage files in Google Drive, all through Telegram. The video starts with a live demo showing how the agent renames uploaded files, combines images (e.g., a selfie with a granola bag), and edits images (e.g., placing granola in front of the Eiffel Tower). The workflow uses a main agent with five tools: two for image generation (combine and edit) and three for file handling (rename, search raw files, search AI images). The creator explains the system prompt, the use of GPT-5.1 with Sonnet 3.5 as fallback, and simple memory via Telegram chat ID. The file handling tools are straightforward, while the image tools are custom sub-workflows that download images, upload them to a public URL service (ImageBB), and call FAL AI to run Nano Banana. The video covers pricing (about 4 cents per image) and suggests improvements like adding a dedicated prompt-optimizing agent, logging to Google Sheets, and integrating with video generation. All resources are offered free in the creator’s Skool community.

194 words

Critical Evaluation

Value of the Information & Strength of the Argument

The video provides a practical, hands-on demonstration of building an AI agent with n8n, showing real-time interactions and results. The argumentation is based on the creator’s experience and the visible functionality of the workflow. The value lies in the actionable steps and the free resources provided, enabling viewers to replicate the system. However, the technical depth is limited; the creator glosses over some details (e.g., exact API configurations) and does not discuss potential limitations or alternative approaches in depth.

Scientific Rigor, Source Quality, Title Accuracy

The video is a tutorial, not a scientific presentation, so it does not cite formal sources. The creator mentions FAL AI and ImageBB as services used, and references his own community for resources. The title accurately describes the content. The video is well-structured with clear timestamps, but the lack of external references and the reliance on personal experience reduce its scientific rigor. The comments show a generally positive reception, with some users asking for clarifications on use cases and technical details.

175 words

Title / Content Match

The title accurately reflects the content, which demonstrates building a no-code Photoshop AI agent using n8n and Nano Banana.

Quality & Reliability

7/10

The video provides a clear, step-by-step tutorial with live demonstrations, but lacks in-depth technical explanations and relies on the creator's experience rather than formal sources.

Chapters

Cited Sources

Concurring Sources

  • n8n documentation — Official n8n documentation, consistent with the workflow described.
  • FAL AI — The service used for image generation, as mentioned in the video.

Contribution & Novelties

The video presents a novel integration of n8n with Google’s Nano Banana model, showcasing a no-code approach to building an AI agent for image editing and management. The main contribution is the modular design using custom sub-workflows as tools, which allows for easy reuse and extension. The tutorial provides a practical template that viewers can download and customize.

Pour aller plus loin :

  • n8n documentation — Official documentation for n8n, useful for understanding workflow automation.
  • FAL AI — The service used for image generation, offering various models including Nano Banana.
  • Google Nano Banana — Information on Google’s image generation model, though the exact page may not be specific to Nano Banana.
  • ImageBB — Free image hosting service used to obtain public URLs for images.

124 words

Radar Profile

The radar profile shows high scores in quantity and quality of information, reflecting the detailed tutorial content. The technical level is moderate, suitable for intermediate users. The reliability is good, but not perfect due to the lack of formal sources. Overall, the video is a valuable resource for those interested in no-code AI automation.

Reliability 7/10

💬 Positif. Sur les 30 commentaires analysés, la majorité exprime de l'appréciation et de la gratitude pour le contenu gratuit, avec quelques questions techniques et demandes de clarification sur l'utilité et la configuration.