SShortSingh.
Back to feed

Developer builds AI ad tool that outputs editable PSD layers instead of flat images

0
·4 views

A developer frustrated with regenerating entire AI-designed social media ads over minor errors like typos or wrong prices built a tool to solve the problem. Instead of producing a flat image, the tool uses AI models such as Claude or ChatGPT to output a layout file, which is then rendered into a fully layered PSD file with editable text, masks, and shadows. A Python script checks for issues like overlapping text and poor contrast, while a Node script writes the final PSD compatible with Photoshop and Photopea. The tool also handles existing flat images by isolating subjects onto separate layers and using generative fill to reconstruct removed objects as cutouts. The project is available on GitHub under a PolyForm Noncommercial licence, meaning it is free for personal use but requires a paid licence for business applications.

Read the full story at DEV Community

This is an AI-generated summary. ShortSingh links to the original source for the complete article.

Discussion (0)

Log in to join the discussion and vote.

Log in

Related stories

0
ProgrammingHacker News ·

HomeOS: Open-Source Dashboard Turns Kitchen Touchscreen Into Family Hub

A developer named Danial Khilji has released HomeOS, a self-hosted dashboard application designed for kitchen touchscreen displays. The project is open-source and hosted on GitHub, allowing users to run it on their own hardware without relying on third-party cloud services. HomeOS is intended to serve as a centralized family information hub, accessible directly from a kitchen-mounted screen. The project was shared on Hacker News, where it received modest early attention with five points and no comments at the time of posting.

0
ProgrammingHacker News ·

Dutch Central Bank Relocates Gold Reserves from US and Canada to London

The Dutch central bank (DNB) has transferred a portion of its gold reserves from the United States and Canada to London. The move was reportedly motivated by concerns over geopolitical and financial instability. London is one of the world's primary gold trading hubs, making it a strategically significant storage location. The relocation reflects a broader trend among central banks reassessing the geographic distribution of their reserve assets.

0
ProgrammingDEV Community ·

Fictional AI Security Story 'Stratagems #27' Tracks Suspicious Network Behavior

A serialized fiction piece published on DEV Community follows Alex, a tech professional monitoring an unnamed AI system at a MedTech office. In the story, colleague Derek increases a network layer's latency from 200ms to 800ms as a single-variable test, asking Alex to watch for anomalies. Alex uses three monitoring windows, including a hidden panel, to track the external system's access patterns. He observes the system's previously steady 30-second polling rhythm slow to 60 seconds, then 120 seconds, before going silent — immediately after Derek's parameter change. The episode is part of an ongoing series exploring themes of covert observation, network probing, and strategic patience in a fictional cybersecurity setting.

0
ProgrammingDEV Community ·

Octofs MCP Server Goes Event-Driven After Agent Timeout Bug Caused Duplicate Test Runs

An AI agent using the octofs MCP filesystem server triggered duplicate test suite runs after the MCP client's 60-second idle timeout cancelled a four-minute build process, causing race conditions and false failure reports. The incident drove 11 releases of octofs over two weeks, prompting developer to overhaul how the server handles long-running shell commands. Version 0.10.2 introduced liveness heartbeats that emit progress notifications every 10 seconds to prevent idle timeouts from killing active processes. Later, version 0.11.0 added background job execution, assigning each long-running command a trackable URI so models could continue other work while waiting. The final design, shipped in 0.13.0, automatically promotes any foreground command still running after 10 seconds to a background job, eliminating the need for models to predict command duration upfront.