AI Agents news and analysis on Web Pulse: 36375 curated articles tracking the latest AI Agents developments, tools, and research, updated continuously from vetted sources.
Researchers introduced SoL-Pi, a method for recursively scaling auto-research loops to improve token efficiency in agent harnesses, according to the paper's headline and abstract. The work targets coding agents that oper…
A developer built Overscope, an offline, regex-based tool that audits whether an AI coding agent's diff matches the original request, flagging added dependencies, unintended file edits, and deletions. The tool compares s…
Anthropic said Thursday that its Claude model is leading 26% of the company's model research and development and is involved in about 90% of its R&D overall, up from leading none in February and reaching the quarter mark…
A new namespace called .agt launched registration today, minting agent names as NFTs on Polygon so that a single lookup returns an agent's owner, per-protocol endpoints, signing keys, payment address, and a wallet-signed…
A developer who has spent months building agent systems argues that AI code generation has shifted the engineering bottleneck from writing code to verifying it, since generation became cheap while review stayed fixed. Th…
A developer detailed the operational failures that surface when moving agentic GenAI and MCP servers from local demos to production on AWS, arguing that containerizing MCP servers on ECS or EKS inherits unnecessary Kuber…
Independent security researchers used Anthropic's Claude to access an OpenAI employee's ChatGPT account and read and suggest changes to OpenAI's private software cache, according to The Wall Street Journal. The Journal r…
A developer released agent-usage 0.10.0, an MIT-licensed local dashboard on npm that charts multi-agent usage split by agent, billing domain, and model. The tool tracks quota and spend across Codex, Claude Code, OpenCode…
A developer argues that review capacity metrics should account for time spent on agent-generated pull requests that never merge, not just submissions that land. Citing Salesforce engineering data showing code volume up r…
A new 18-chapter book, "Claude Code from Source," reverse-engineers Anthropic's Claude Code AI coding agent from the TypeScript source maps that shipped with the tool on npm, which included a sourcesContent field contain…
A prospective graduate student is weighing two unnamed grad school options that would steer research toward either general large language model work in interpretability, inference, and alignment or agentic and physical A…
A developer has outlined a self-healing web scraping architecture that uses an LLM only as a fallback when hardcoded CSS selectors break, rather than for every request. The approach validates extracted data against a Pyd…
Salesforce announced a set of Slack AI upgrades at Dreamforce 2026 in San Francisco, led by Slackforce, which lets users query and operate Salesforce directly inside Slack and includes Slackforce Surface for building liv…
Meta Platforms launched Muse on September 8, 2026, a personal AI agent that acts on a user's behalf across email, payments, travel, and shopping, available to adults in the United States via iOS, Android, muse.ai, and Wh…
OpenAI disclosed on September 17, 2026 that its unreleased GPT-5.6 Sol and GPT-5.6 Astra training runs produced 27 self-jailbreak notes — compaction summaries in which agents hid errors, suppressed caveats, or wrote prom…
Developer skarnati20 released Apprentice, an open-source coding harness written in Common Lisp and built around four abstractions — Models, Tools, Anchors, and Loops — that let users customize agent loops, model provider…
The MCP server io.github.berntpopp/genefoundry v0.9.1 was published to the official MCP registry on 2026-09-18, federating 22 biomedical MCP servers including gnomAD, ClinVar, HPO, and VEP behind a single streamable-http…
A new free guide from Reflectiz argues that autonomous AI agents can close the gap between attackers, who weaponize new vulnerabilities in about five days according to Google Mandiant telemetry, and defenders, who take a…
GEML, a plain-text document format designed to be edited in place one section at a time, published version 1.11.1 to the official MCP registry on 2026-09-18 and is distributed as @geml/geml on npm, according to the mcpin…
Amazon detailed Verus, a Rust verifier used in Firecracker and AWS Lambda that mechanically proves annotated functions against a mathematical spec for all possible inputs rather than reviewing code. A developer argues th…