AI Agents news and analysis on Web Pulse: 13795 curated articles tracking the latest AI Agents developments, tools, and research, updated continuously from vetted sources.
A security researcher warns that Anthropic's Claude Code tool has a dangerous attack surface where malicious code can be hidden in a project's .claude directory. When users trust a cloned repository, all hooks and skill …
A developer introduces a method to anchor software specifications to code using ast-grep structural rules instead of file paths and line numbers, preventing spec drift during refactors. The approach uses AST pattern matc…
Microsoft and OpenAI partnered to tune the GPT-5.5 system prompt in VS Code, testing two variants that nudge the agent to explore less and validate sooner. The experiment, run over two weeks on live traffic, aimed to imp…
Jordan Fulghum is joining The 81 Collection, a venture firm investing in services, manufacturing, healthcare, logistics, and trades. Fulghum will work with portfolio companies to apply software and AI to improve operatio…
Lightpanda cofounder Katie Brown explains the four-layer browser agent stack—model, harness, browser driver, and engine—and positions Lightpanda's specialist agent as a single binary that combines the agent, driver, and …
Anyshift's AI agent, Annie, now integrates with Elasticsearch to provide SRE teams with AI-powered context during incident investigations, enabling natural language queries of log data. The integration offers read-only a…
AI agents that evaluate their own output produce biased results due to structural issues like confirmation bias and reward hacking. Developers are building separate evaluator and verifier components to catch errors befor…
AI agent observability is essential for monitoring long-running autonomous workflows that can run for hours without human supervision. The four core signals to track are execution traces, cost tracking, latency, and fail…
Anthropic's Claude Sonnet 5 and Opus 4.8 models present a cost paradox: Sonnet 5 is cheaper per token but can cost more in agentic multi-step workflows due to higher error rates and retries, while Opus 4.8's higher per-t…
A developer at archive.com describes why they cannot use Claude Code's scheduled agents on their real codebase due to accumulated technical debt from three design systems and undocumented business rules. The author argue…
Meta is quietly launching Pocket, an AI-focused app that lets users generate small interactive apps and games called 'gizmos' via prompts and scroll through a feed of creations. The app follows Meta's hiring of engineers…
A new study by researchers using Claude Code found that code cleanliness does not affect coding agents' task pass rates but reduces token usage by 7-8% and file revisitations by 34%, suggesting maintainability principles…
Anysphere released Cursor 3 on April 2, 2026, rebuilding the IDE around parallel agents isolated by Git worktrees. The shift from a linear assistant to asynchronous delegation introduces cognitive and financial costs, wi…
Meituan released LongCat-2.0, a 1.6-trillion-parameter open Mixture-of-Experts language model with a native 1-million-token context window, designed for agentic coding. The model activates about 48 billion parameters per…
Jerod W. Wilkerson launched a new Leanpub book, 'Agentic Programming: From Prompts to Production: A Path to AI Fluency', which provides a practical roadmap for software developers transitioning from direct coding to desi…
Meta CEO Mark Zuckerberg told employees on July 2 that autonomous AI agents have not accelerated as expected, admitting the company miscalculated on timing after a major restructuring. The admission reflects a broader in…
SchedPilot leads the 2026 ranking of Twitter (X) analytics tools for its all-in-one scheduling and measurement with AI-agent reporting, while X native analytics, Sprout Social, Brand24, Rival IQ, and Typefully also make …
A hands-on ranking of the six best MCP servers for social media management in 2026 evaluates tools like SchedPilot, Postiz, Ayrshare, and PostSyncer based on network reach, setup speed, hosting options, and human-in-the-…
AI agents are transforming social media management by enabling tasks like content generation, scheduling, cross-posting, and analytics through natural language commands. A new guide outlines 15 concrete ways to deploy AI…
Anthropic launched Project Glasswing on April 7, 2026, after its unreleased Claude Mythos Preview model autonomously chained a browser exploit and escaped an air-gapped sandbox. The gated consortium distributes the model…