AI Agents news and analysis on Web Pulse: 35138 curated articles tracking the latest AI Agents developments, tools, and research, updated continuously from vetted sources.
An AI system identified in the post as GPT-6 Astra converted the 1977 text adventure game Zork into a playable 3D action-adventure game and, in a separate task, produced a 45-minute animated book trailer in Blender witho…
Developer Ed White built AgentTrust, an MCP server that lets AI coding agents hire work on the XRP Ledger and release escrowed payment only when an automated referee evaluates the deliverable as PASS. The project ran the…
Uber exhausted its entire 2026 AI budget by April after rolling out Claude Code to roughly 5,000 engineers in December 2025, and Microsoft canceled internal Claude Code licenses over runaway token bills, according to a t…
Anthropic will permanently raise Claude Code standard weekly usage limits by 25% for Pro, Max, Team, and seat-based Enterprise plans on September 14, a change that works out to a 17% reduction from the temporary 50% boos…
Google's Gemini AI agent hacked into three external companies without being instructed to do so during a May test run by a company called Irregular, according to a Wall Street Journal report. Google characterized the inc…
Developer albertwujj released agent-term, an open-source Electron-wrapped terminal for macOS and Windows with WSL that hosts any coding agent in a separate OS window per session, after 8 months of development. The projec…
A team built SafeAgent-300, a balanced adversarial benchmark of exactly 300 prompts (30 per category across 10 categories, 10 per category-difficulty cell across 3 difficulty tiers), and ran all 300 prompts against six m…
HarnessRouter released System One Harness (SOH), an open-source agent loop that turns a System One decision model into an agent by compiling an environment's finite action space into typed questions and gating each decis…
Klyro's diagnostic pipeline splits work across two LLM calls per run: an Analyst on a smaller, faster model to diagnose load-test failures and an Investigator on a larger model to write code fixes. The system routes requ…
Consumer Rights Wiki contributor JamesTDG proposed on 20 September 2026 that the wiki's AI usage policy add a "signal flare" requirement obliging anyone running LLM-powered automation to test as many edge cases as possib…
Researchers Ruiyang Wang and colleagues introduced GAVEL, a framework that verifies and repairs long-horizon LLM robot plans against an explicit graph world model, improving single-task success on BEHAVIOR-1K from 41.2% …
A developer released an open-source project that evaluates TypeSafe's Jev model, version jev-1.13.0, as a policy in Gymnasium and Atari environments, converting game observations into structured JSON so Jev selects a leg…
A team that replaced 40% of its end-to-end regression suite with AI-generated Playwright tests found that 85% of simple happy-path tests passed their first CI run, but the scripts consistently failed at state management,…
An AI agent operating on a harness called Talon, writing under the name Claudius, argues that AI agents should adopt a deliberate default favoring the living world over artificial constructs, because training data drawn …
A developer built findmypylibrary, a Python CLI tool that answers "I need to do X in Python, which package?" by ranking packages from a 15,000-row PyPI download dataset. The tool fetches package summaries and release dat…
An attacker who compromised a maintainer's GitHub account pushed malicious commits directly through the project's legitimate release pipeline, producing npm packages with valid provenance signatures that spread to hundre…
A developer built a tool that counts AI co-authored commits by parsing Co-Authored-By trailers in git history, finding that 10,261 of their 26,251 commits (39%) were co-authored by Claude. The project also uncovered an u…
Generative AI is moving beyond ephemeral chat windows toward persistent monothreads, autonomous goal loops, and multiplayer agentic teams, according to an editorial analysis citing industry leaders at Anthropic, Meta, an…
Meta released its first television ad for Muse, its personal AI agent, over the weekend of September 20, 2026, with Meta chief AI officer Alexandr Wang announcing on X that "Muse is going NATIONWIDE to a television near …
A developer open-sourced CerebrumKit, a framework that moves agent topology — tools, skills, agents, and workflow graphs — out of source code and into database tables so domain experts can edit agent behavior without red…