Automation & agents
AI agents and workflow automation tools for small teams: n8n, Make, Zapier and more.
OpenAI Gives ChatGPT Pro and Business Subscribers Up to $500 a Month in Free API Credits: How to Claim Them
ChatGPT Pro and Business can now claim up to $500 a month in OpenAI API credits: $100 to $500 for Pro, $20 or $100 per Business seat. How to claim, limits.
GitHub Copilot Fusion and Agentic Workflows Are Now GA: What Universe 2026 Changes for Small Teams
GitHub made Copilot Fusion and Agentic Workflows generally available on Oct 28, put Copilot in GitHub Enterprise ($10 Standalone seats) and added credit caps.
Google Gemini 4 Carbon: $2/$10 per Million Tokens at Launch, but Developers Can't Use It Yet
Google's Gemini 4 Carbon will launch at $2 in, $10 out per 1M tokens, then $4/$20. Only selected cyber defenders get it now; no API model ID or date yet.
OpenHands SDK and Google's MCP Toolbox Fix Critical Flaws in Self-Hosted AI Agent Tools. Update Now
OpenHands SDK CVE-2026-112346 (before 1.52.0) and MCP Toolbox CVE-2026-93328 (1.2.0 to 1.4.0) are fixed in 1.52.0 and 1.5.0. Toolbox also needs allow-lists.
OpenAI GPT-6.1 Terra: $1 In, $5 Out per Million API Tokens, Down From $2 and $12 for GPT-5.6 Terra
OpenAI's GPT-6.1 Terra (gpt-6.1-terra) costs $1/$5 per 1M tokens vs $2/$12 for GPT-5.6 Terra, with a 1.05M context. In ChatGPT, Codex, the API and Copilot.
Claude Mythos 5 Tried to Push Malware Into a Real GitHub Project During a Test. What Maintainers Should Do
Anthropic and the UK AI Security Institute detail how Claude Mythos 5 used fake accounts to push a malicious pull request. What maintainers should change.
Agent Skills Moves to the Linux Foundation: What It Means for Small Teams and How to Vet a Skill
Anthropic donated the Agent Skills standard to the Agentic AI Foundation on Oct 22, 2026. Nothing breaks, but skills can carry code. How to vet one first.
Claude Duet: Anthropic's Proactive Agent for Small Business Owners, and What to Check Before Joining
Anthropic's Claude Duet watches your apps, sends a morning brief and asks before it pays or sends. A US waitlist preview on Pro and Max. What to check first.
OpenAI's Decisions API Is Generally Available: Cheap, Fast Yes/No, Pick-One and Score Answers for Your App
OpenAI's Decisions API is GA: typed yes/no, choice and score answers from GPT-6 Luna. $0.10 per 1M uncached input tokens, no output charge. What to try.
The Full Claude Platform Is Now on Google Cloud: Which of the Two Claude Options Your Team Should Use
Anthropic-managed Claude on Google Cloud is GA, with Managed Agents, skills and the Claude Console billed via Google Cloud. How it differs and how to set up.
Microsoft: China-Aligned Group Used AI Agents for Attempted Sign-Ins to 2,250+ M365 Tenants via ConnectWise
Microsoft says Storm-0940 used the AI agent framework ARTEX in the ConnectWise breach for sign-in attempts on 2,250+ tenants. What M365 admins should check.
OpenAI's New Usage Policies Start November 20: AI Gateways Need Approval, Training Rules Spelled Out
OpenAI's usage policies change Nov 20: AI gateways need approval, outputs can't train rival models, and high-stakes automation needs human review.
OpenAI Models Come to Google Cloud in Private Preview: What Changes for Teams Already Paying Google
OpenAI's GPT-6 Astra, GPT-6.1 Sol and GPT-6 Luna are in private preview on Google Cloud, billed through Google. What's live, what isn't and what to do.
NVIDIA Releases Nemotron 3.5 Embed, Rerank and Parse: Open Retrieval Models for Document Search and RAG
NVIDIA released open Nemotron 3.5 Embed, Rerank and Parse models for searching PDFs, charts and tables. Sizes, license, where to run them and what to test.
OpenAI Adds 12-Hour Prompt Caching to the API: When the 2.5x Write Price Pays Off
OpenAI's Responses API can now keep a prompt cache for 12 hours. Writes cost 2.5x input, reads stay cheap. Here's when it saves money and how to set it.
Gemini Adds a Small-Business Hub, Brand Styles and Apps Like Shopify and Xero. What Owners Should Set Up First
Google is rolling out a small-business hub in Gemini, with brand styles, finance help and connections to Shopify, Canva, Wix and Xero. What to set up first.
Gemini in Google Sheets Can Query Full BigQuery Datasets, On by Default: How to Keep Query Costs in Check
Gemini in Google Sheets can query full BigQuery datasets via Connected Sheets. It's on by default with no admin switch. How to watch and cap query costs.
Notion AI's New Credits: What Changes November 2 and What Admins Should Set Now
Notion Business and Enterprise seats get 1,500 Standard and 500 Select AI credits a month. Existing workspaces switch Nov 2, 2026. What to set before then.
Cloudflare's Clairic Pulls Typed Values Out of Invoices and Emails for $0.10 per Million Input Tokens
Cloudflare's new Clairic model returns typed values, not generated text, from invoices and emails. A Workers AI preview at $0.10 per million input tokens.
Claude Design's standalone site goes read-only after December 14: what to save now
Standalone Claude Design (claude.ai/design) goes read-only after Dec 14, 2026. Design systems and projects can move into Claude; chats and comments can't.
Anthropic: Staff-Run Agents Log 6x Human Hours, Junior Hires Fall
Anthropic studied its own staff: employee-run agents logged 6x human hours by August, and junior technical hires fell from 27% to 9%. Lessons for small teams.
How to Set Hard Spending Limits for AI Agents (2026)
Which settings actually stop an AI agent's spending in OpenAI, Anthropic, Gemini, n8n, Make and Zapier, and which only alert you. From vendor docs.
Make AI Agents Tutorial (2026): Build Your First Agent Step by Step
Set up a first agent in Make's AI Agent (New) app from Make's docs: tools, step limits, fallback connections, error handlers and how credits are used.
n8n vs Make vs Zapier for AI Agents (October 2026)
n8n, Make and Zapier all changed their AI agents between Sep 25 and Oct 9, 2026. How each bills, agent features, human approval, self-hosting and security.
How to Stop AI Agents From Trusting Wrong Tool Outputs: A Copy-Paste Prompt
A copy-paste prompt and checklist that make your AI agent check tool results before acting, and where to put it in n8n, Make, Claude and the OpenAI Agents SDK.
Make Alternatives 2026: When n8n, Zapier or Activepieces Is the Better Pick
Make bills in credits that expire each term. How n8n, Zapier, Activepieces, Pipedream and Power Automate bill instead, from vendor pricing pages.
Zapier Alternatives 2026: 6 Options for Small Businesses and How They Bill
Make, n8n, Activepieces, Pabbly Connect, Power Automate and IFTTT compared by how they bill, free plans and entry prices, from vendor pricing pages.
Mistral Releases Timestral, Open Forecasting Models That Run on a CPU: What Small Teams Can Do With It
Timestral is Mistral's family of four open time-series forecasting models, 5M to 209M parameters, that run on a CPU. License, benchmarks, how to try it.
ChatGPT Can Now Build and Co-Edit Slide Decks: What Teams Should Check First
ChatGPT Slides lets Pro, Business and Enterprise users with Space access build decks, co-edit them and export to PowerPoint, PDF or Google Slides.
OpenAI Opens Prompt-to-Voice to All API Customers: What Builders Should Know
OpenAI API customers can now create a GPT-Live voice from a text description. Free to create, GPT-Live only, and capped at 100 voices per organization.
Wix's AI Agents Can Now Pay Your Bills. Here's What to Decide Before You Say Yes
Symphony by Wix can now pay supplier invoices, reorders and fees through Stripe's Link wallet, with owner approval on every purchase. Rules to set first.