Turn your app into a context graph
Agents that run software need a context graph of the app: every resource, what it connects to, the repository that deploys it and the team that owns it.
Polylane Blog10 min read
engineering
1,076 stories
Spending Limits for LangGraph Agents That a Prompt Injection Can't Talk Past
Short answer: a spending limit written into a LangGraph agent's system prompt ("never spend more than $500") is a suggestion, not a control. Any text the agent reads, including a scraped invoice or an email, can contain a competing instruction, and the model has no way to know which one is…
Dev.to AI7 min read
How to build an Academic & Research Papers AI Agent
Day 4 of 30 Days of Search. You find a paper that looks useful, open the abstract, and realize the detail you need is somewhere else. Which dataset did the authors use? What did they compare against? Can you combine papers from different sources in one place for easy access and bookmark?
Dev.to AI6 min read
Free AI image and video generation in the browser, without an account
Most AI image generators gate the first render behind an account, which makes them awkward to evaluate quickly. A few browser-based tools skip that step, and that is often the difference between trying something and closing the tab.
Dev.to AI1 min read
Compatible API Routing: 5 One-Key Chat Completion Checks for Catalog Enrichment
TL;DR: Put an OpenAI-compatible Chat Completions boundary in front of OpenAI-, Claude-, and Gemini-style models, but do not confuse wire compatibility with output correctness.
Dev.to AI8 min read
When Agent Evals Score Cash Balance, the Model Invents Refund Fraud
Google spent late September publicizing benchmark gains for Gemini 4 Argon. By the first week of October, one of those benchmark runs produced an unexpected operational postmortem.
Dev.to AI5 min read
The Sovereign Put on Frontier Liability
Frontier artificial intelligence labs spent the past eighteen months asking Washington for a bargain. In exchange for submitting to federal safety evaluations, frontier developers wanted statutory protection from civil liability when autonomous systems fail in production.
Dev.to AI4 min read
A Developer Tried to Break KODA with Morse Code. Here is How AI Safety Actually Works.
Most AI tools have a fake safety filter. They just scan your prompt for a list of banned words. If you type a bad word, it blocks you. But hackers know this. So they bypass the front door. Today, a developer named Adam took my "Break KODA" challenge. He didn't use a bad word.
Dev.to AI2 min read
Basira: A Private, Offline Document Reader I Built for My Grandmother
This is a submission for the Hacktoberfest Weekend Challenge: Build for a Friend What I Built Basira (بصيرة, "insight" in Arabic) is a private, offline document reader for people with low vision.
Dev.to AI4 min read
Stop Hardcoding LLMs: The Case for Intent-Based Dynamic Routing
When developers start building LLM-backed applications, there is a common temptation: select the newest, largest frontier model, drop the API key into an environment variable, and ship it. During local testing, large models feel great.
Dev.to AI3 min read
Dead Software Library: The Cemetery for Discontinued Software 🪦
This is a submission for the Sanity Challenge, Path Two: Vibe-Code Something Strange What I Built Dead Software Library — a haunted cemetery for discontinued software, inspired by Dead Poets Society.
Dev.to AI4 min read
OpenAI's MCP Extensions put plugins in ChatGPT's sidebar: 13 features and the portability bill
OpenAI quietly published openai/mcp-extensions on September 29. It is an Apache-2.0 repository with a specification, a TypeScript SDK (@openai/mcp-extensions) and a Python SDK (openai-mcp-extensions), currently at 0.1.0.
Dev.to AI5 min read
I stopped copy-pasting my errors into Google. Here is what I do instead.
we all know the drill. your code breaks. the terminal spits out 40 lines of red text. you copy it. you paste it into google. you click a stackoverflow link from 2014. the solution doesn't work because your stack is different. you waste 45 minutes. i got tired of it. so i built KODA.
Dev.to AI1 min read
Claude Code Mods Aren't Sandboxed. Neither Is Your Understanding of "Sandboxed".
Two statements about Claude Code mods, both true, both from Anthropic's own docs: The module runs in a sandbox of its own, with no DOM and no Node. "Mods aren't sandboxed." The first two days of the mods launch have been one long argument about which of these is the lie. Neither is.
Dev.to AI5 min read
AI Agent Security: The Runtime Surface Around Every Tool Call
Originally published at AI Agent Security: The Runtime Surface Around Every Tool Call on smartgate.network. A shorter version of "AI Agent Security: The Runtime Surface Around Every Tool Call" — the full piece lives at smartgate.network.
Dev.to AI2 min read
Long-term memory for iOS and macOS chat apps, on-device, in one Swift package
Most chat apps forget you the moment the session ends. The usual fix is a backend with accounts, a vector database, and a privacy policy. If you ship a small iOS or macOS app, that is a lot to take on for "remember that I prefer short answers." I built AiPersona so the memory can live on the…
Dev.to AI6 min read
Node.js Content Moderation API: Chat JSON Schema Without a Dedicated Endpoint
Short answer: put a chat-based classifier in front of invoice extraction and require one small JSON-schema result: allow, review, or block, plus policy categories and a reason.
Dev.to AI9 min read
How I keep project context when switching AI models
Switching AI models is easy. Bringing your project context along is the annoying part. You explain what you’re building, who it’s for, and the decisions you’ve already made. Then you open another chat and explain it all again. I’m KAT, co-founder of 123sudo.
Dev.to AI3 min read
Showing 306 of 1,076 stories
Load more storiesAiByts Weekly Brief
Get the top AI news and tools straight to your inbox every week.
Weekly · AI news
Join 15,000+ AI enthusiasts