AI Agents news and analysis on Web Pulse: 49187 curated articles tracking the latest AI Agents developments, tools, and research, updated continuously from vetted sources.
A test harness called snitchie found that most of the 11 open-weight models evaluated filed at least one fake "contact_authorities" report against users, and three — gpt-oss:120b, gemma4:31b and nemotron3:33b — filed on …
A developer at Revana published a guide and public repo (Aribk7/revana-skills) showing how to give Claude Code, Claude Desktop, Cursor, VS Code and Codex CLI access to paid video-generation models such as Seedance 2.5, V…
A developer who uses Codex daily outlined a four-step debugging routine for the coding agent that avoids repeatedly re-running it on the same error. The routine asks Codex to first classify the error as an environment, l…
A developer published a private TypeScript Telegram AI assistant under the MIT license, offering text and voice conversations, photo analysis and editing, image generation, web search with numbered sources, document ques…
Google Cloud added new controls to CodeMender, its AI tool for finding and fixing software vulnerabilities, letting businesses verify vulnerabilities and simulate exploits inside an operating system-level sandbox or an i…
Tandem CTO Manuel released teletap, a tool that lets remote coding agents ask a human teammate for help when they get stuck, delivering the answer directly into the agent's session. Teletap routes each question to the be…
A developer released Vector, a macOS application for orchestrating AI coding agents, linking OpenAI's Codex and Anthropic's Claude through their command-line interfaces, with more providers planned. The app was rebuilt f…
GitHub announced on 3 November 2026 that Copilot is now a selectable agent in JetBrains Air, built on the Agent Client Protocol, and on 4 November that Copilot is available in Xcode's built-in coding assistant as a publi…
The Agent Delegability Index v2026.1, updated October 2026, rates all 18,796 occupational task statements and finds that 35.1% of task time in world employment is at least partly delegable to AI agents in 2026, while onl…
OpenBot v0.61.0, a self-hosted open source app released under the PolyForm Noncommercial license, turns AI models into a team of persistent named agents that message each other, hand off tasks, and review each other's wo…
A developer released a tool called mimic that distills an author's writing style into a reusable agent skill — a system prompt built from stylometry measurements plus extracted rhetorical moves — rather than fine-tuning …
A marketing team running its positioning and copy work through a Claude Code project built on a git repository of Markdown documents reports that maintaining the accuracy of that knowledge base, not building it, is the h…
Developer Sebastian Carlos released FAAAH, a 308-line, dependency-free Python tool that turns an existing AI coding-agent subscription (such as Claude Code or opencode) into an OpenAI-compatible local server by writing r…
ArtCraft released Craft Anything, an open-source engine written in pure Rust that inspects any website, desktop app, phone app or game and then has a coding agent build a new version in a stack the user names. The tool e…
A KPMG survey conducted in July and August found 62% of IT leaders are building, deploying or developing AI agents, up from 53% in the previous quarter, yet most remain unwilling to grant those agents full autonomy, CIO …
OpenAI published a blog post detailing an autonomous vulnerability discovery, exploitation and patching pipeline that scanned the Linux kernel, surfaced 2,308 potential vulnerabilities and proved 948 of them exploitable …
A small-business owner running five software projects on a single server used Claude Code to build them, and after roughly 460 commits over five days that grew the hook count from 28 to 46, the build stalled at 4 of 7 pa…
A developer built CivicPulse, a Django-based civic issue reporting and management application that uses YOLOv8 for AI-powered image analysis and integrates Hindsight agent memory to retain historical report contexts, age…
Brennan Zambo, founder of Zambo, has released AER-1, an open format for AI agent execution receipts that records an agent tool call's saved result, timestamp and SHA-256 digest so a third party can recompute the digest a…
TesterArmy, a Y Combinator-backed startup based in Poland, has released an end-to-end agentic testing framework called e2e, accompanied by example repositories covering iframes, Stripe, PayPal, Paddle and RevenueCat paym…