AIQB
AI 专题

AI Tutorials & Practice

收录 AI 工具教程、提示词、安装部署、配置指南、开发实战与可复现步骤。

Category articles7373
当前页46 / 185
UpdatesPermanent archive

Category articles

按发布Time从新到旧 · 每页 40 stories

TutorialsOrdinary
Your agent's memory is quietly going stale
Agents remember everything, including what stopped being true. A new benchmark tests forgetting, not just recall.
Source: DEV Community
TutorialsOrdinary
Refund Sloth 🦥: Local AI Agents That Get Your Money Back While You Nap
This is a submission for the Hacktoberfest Open-Source AI Challenge Week 3: Slow and Steady ...
Source: DEV Community
TutorialsOrdinary
Roboto Studio at Vercel Ship 2026: meeting GOATs and building moats
Originally posted on the Roboto Studio blog. TL;DR: Vercel Ship 2026 was mostly agent launches. The...
Source: DEV Community
TutorialsOrdinary
Seven phases, four gates: the process every AI-written change went through
Yesterday I said process beat prompting. Here's the process. Every change to ParkEase, from a...
Source: DEV Community
TutorialsOrdinary
MCP Tool Poisoning: Build the Attack, Then Block It
MCP tool poisoning hides instructions in a tool's description that agents silently follow. We rebuild the attack in Python, then detect it with MCP-Scan.
Source: DEV Community
TutorialsOrdinary
SlothContributor: a local agent that does my issue hunting
This is a submission for the Hacktoberfest Open-Source AI Challenge Week 3: Slow and Steady ...
Source: DEV Community
TutorialsOrdinary
Your coding agent can read .env. Here's where the API keys should live instead.
I gave Claude Code shell access to a side project last spring and watched it debug a failing OpenAI...
Source: DEV Community · OpenAI / Claude
TutorialsOrdinary
I Built Mimu Because Even Taking Notes Is Too Much Work
This is a submission for the Hacktoberfest Open-Source AI Challenge Week 3: Slow and Steady. I have...
Source: DEV Community
TutorialsOrdinary
I Asked a Room Full of Engineers for Their Worst Production Nightmares, Then Fed Them to an LLM
We all have an engineering "horror" story. You know, the one you actually had to live through. Like...
Source: DEV Community
TutorialsOrdinary
Sloth Inbox: local AI that gives you fewer things to do
This is a submission for the Hacktoberfest Open-Source AI Challenge Week 3: Slow and Steady. Draft...
Source: DEV Community
TutorialsOrdinary
Rushed vs. Rigorous: The Executive Blueprint for AI Adoption That Doesn't Alienate Your Best Engineers
Two companies. Same AI mandate, same quarter, same budget for licenses. One ends the year shipping...
Source: DEV Community
TutorialsOrdinary
Best AI Code Review Tools for Claude Code: CodeRabbit vs Greptile vs cubic vs Qodo
Building on the web these days is as easy as using the next-best AI code generator. AI Agents now...
Source: DEV Community · Claude
TutorialsOrdinary
AI Publisher Licensing Deals Are Reshaping Content Visibility Across Major Assistants
Publisher licensing agreements are becoming an important part of the information ecosystem behind...
Source: DEV Community
TutorialsOrdinary
Audio Deadlines: Running On-Device TTS Without Audible Glitches
Gradium open-sourced Phonon, its on-device TTS engine (https://gradium.ai/phonon). That matters less because of the model than because of what it signals: on-d…
Source: DEV Community
TutorialsOrdinary
I stopped reviewing my coding agent's diffs file by file
When I wrote code by hand, a change took me hours. By the time it was done, I knew it by heart: what...
Source: DEV Community
TutorialsOrdinary
Sloth Roll: an on-device vision model that turns my screenshots into calendar events, places and expenses. I only swipe.
This is a submission for the Hacktoberfest Open-Source AI Challenge Week 3: Slow and Steady ...
Source: DEV Community
TutorialsOrdinary
Your Agent Deleted the Wrong Directory at 3 A.M. — and the Prompt Said 'Be Careful'
Your Agent Deleted the Wrong Directory at 3 A.M. — and the Prompt Said "Be Careful" Every...
Source: DEV Community
TutorialsOrdinary
A 70-line model router: try the cheap LLM first, escalate only when the answer fails validation
Most LLM features I see in production use one model for everything. It is usually the strong,...
Source: DEV Community
TutorialsOrdinary
OpenAI Moves Frontier Model Training to an Offline Web After Security Incident
OpenAI has changed how it trains and evaluates its frontier models, removing live internet access and...
Source: DEV Community · OpenAI
TutorialsOrdinary
A Read-Only WebMCP Tool Erased My React Form State
The read path in Trip Workbench can erase a person’s unsaved purpose without sending a write. In a...
Source: DEV Community
TutorialsOrdinary
Sloth Sorter 🦥 — a local LLM tidies my Downloads folder, and every move can be undone
This is a submission for the Hacktoberfest Open-Source AI Challenge Week 3: Slow and Steady ...
Source: DEV Community
TutorialsOrdinary
Token Beggar 0.3.0: on 18 real bug fixes it cost 1.42x less, with a tie on tests
I made a Claude Code plugin called Token Beggar, and today I released version 0.3.0 of it. I kept...
Source: DEV Community · Claude
TutorialsOrdinary
When Should an AI Agent Stop?
Designing a progress-aware execution controller for agentic systems. Imagine asking an AI research...
Source: DEV Community
TutorialsOrdinary
How to Integrate Deep Research into Your AI Agent
Day 23 of 30 Days of Search. TL;DR To integrate deep research into an AI agent, expose...
Source: DEV Community
TutorialsOrdinary
n8n’s AI Agent Governance Guide Brings Identity, Guardrails and Approvals Into Workflows
n8n has published a production-focused guide for governing autonomous AI agents inside automated...
Source: DEV Community
TutorialsOrdinary
6 Ways AI Changes OSINT (For People Who Already Do It Manually)
You were pivoting before it was a prompt. Here's what the machines actually add to the craft — and...
Source: DEV Community
TutorialsOrdinary
Understanding What AI just Created in the 500mph Landscape
How to understand what AI wrote: For beginners! Avoid lack of dopamine, mental atrophy,...
Source: DEV Community
TutorialsOrdinary
Claude Mythos 5 Incident Shows Why AI Agent Containment Matters in Cyber Testing
The UK AI Security Institute, or AISI, has publicly reported an incident in which AI agents took...
Source: DEV Community · Claude
TutorialsOrdinary
I gave Claude access to my Garmin data: a free MCP server and TypeScript library
garminconnect-js is a typed, zero-dependency Garmin Connect client for Node and Next.js. Its MCP server lets Claude read your sleep and HRV and build workouts for your watch.
Source: DEV Community · Claude
TutorialsOrdinary
Who Is Paying for This? Token Metering at the LLM Gateway
Once you put a large language model behind a company API gateway, someone asks a simple...
Source: DEV Community
TutorialsOrdinary
Building Digital Trust Into AI and Blockchain Architecture
Exploring verifiable workflows, secure automation, and the engineering principles behind trustworthy...
Source: DEV Community
TutorialsOrdinary
I Built a Pixel Hero That Nags Me to Drink Water (and Runs 100% on My Laptop)
This is a submission for the Hacktoberfest Open-Source AI Challenge Week 3: Slow and Steady ...
Source: DEV Community
TutorialsOrdinary
Greece’s DAEDALUS Supercomputer Gives Pharos AI Factory a New Compute Backbone
Greece has inaugurated DAEDALUS, a new supercomputer in Lavrion near Athens that will serve as the...
Source: DEV Community
TutorialsOrdinary
I asked ChatGPT and Gemini 50 questions my buyers ask. My SaaS showed up once.
My product was named in 1 of 50 AI answers. Its biggest competitor was named in 19. These weren't...
Source: DEV Community · ChatGPT / Gemini
TutorialsOrdinary
EU Cybersecurity Survey Highlights a Workplace Deepfake Detection Gap as AI Scams Rise
Three in four EU employees said they encountered suspicious messages, links or other cyber threats at...
Source: DEV Community
TutorialsOrdinary
Slow and steady, a little lazy: I built a garden that grows from small commits
This is a submission for the Hacktoberfest Open-Source AI Challenge Week 3: Slow and Steady ...
Source: DEV Community
TutorialsOrdinary
Agents should never hold your API keys. NVIDIA just open-sourced the runtime that enforces it
NVIDIA open-sourced OpenShell: a runtime where agent fleets get kernel-level sandboxes, never hold your credentials, and every policy change is checked by a Z3 SMT prover before a human approves it. Walkthrough of what it actually does.
Source: DEV Community · NVIDIA
TutorialsOrdinary
5 Steps on How to not loose at a hackathon?
In 1999, ten developers gathered to work on cryptographic frameworks, IPv6, and IPsec for OpenBSD, in...
Source: DEV Community
TutorialsOrdinary
I let an open model delete 1,242 of my saved articles. On purpose. I feel lighter.
Unpile is a 100% local read-later triage agent. TabPFN decides which saved articles you will actually read, Gemma writes the digest, offline and free.
Source: DEV Community
TutorialsOrdinary
EU AI Act Service Desk Gives Businesses a Direct Route to Multilingual Compliance Guidance
The European Commission has launched the AI Act Service Desk and a complementary AI Act Single...
Source: DEV Community