Full-text search across 66089 articles. Combine with topic and date filters; results sorted by relevance.
Otōto released version 2026.15 beta5 in October 2026, changing its ask feature so Claude no longer receives the whole task including the fix and tests, and instead is told that ask reports what the code does while decidi…
Runta's FrontierHarness v1.0 benchmark, evaluating nine harnesses across 12 configurations on software development and terminal tasks, found pass rates ranging from 50.0% to 66.7% but cost per completed task varying from…
Security researcher Boschko published a technical breakdown of UniBLEed, a wormable exploit chain that achieves unauthenticated root remote code execution on any Unitree G1 humanoid robot within Bluetooth range, bypassin…
Cloudflare launched Dynamic Workers, a new primitive for spinning up isolated Cloudflare Workers with a single call, enabling safe execution of arbitrary user code. The feature addresses limitations of eval by providing …
Amazon Web Services (AWS) published a technical guide on designing custom reward functions for multi-turn reinforcement learning (RL) with Amazon Nova Forge, emphasizing that the reward function determines what the model…
An engineer tested Anthropic's Claude products—Claude Code, Claude Desktop, and Claude Cowork—and confirmed that all three use Client ID Metadata Document (CIMD) on user-initiated OAuth flows, contrary to some third-part…
Boris Cherny, creator of Claude Code at Anthropic, revealed at a private WorkOS event on June 2, 2026, that he no longer prompts Claude directly but instead writes loops that prompt the agent. Peter Steinberger of OpenAI…
SpaceX acquired AI coding startup Anysphere, maker of Cursor, for $60 billion in stock days after its blockbuster IPO, signaling that AI developer tools have become strategic infrastructure. The deal, which includes a $3…
MSFM (MindStudio-Flavored Markdown) adds three annotation primitives — block annotations, inline annotations, and pointers — to standard Markdown, enabling the format to compile natural language specifications into full-…
Non-coders are using AI programs to create custom apps for specific personal needs, a trend known as "vibe coding" that has gained traction since late 2025. Account manager Shayan Mirzazadeh, who failed computer science …
Vercel added a new value, "connect-branding-updated", to the `name` field of its `ListEventType` object, reachable from one Vercel operation, which the company labeled an additive change. The change breaks exhaustive swi…
OWASP's Q3 2026 exploit roundup identifies three prompt injection patterns that continue to work in the field — instructions hidden in web pages and links, prompt execution, and poisoned agent memory — and OWASP kept pro…
Vercel added four new values to the `name` field on its `ListEventType` object, which is reachable from 1 Vercel operation: `project-git-credential-attestation-created`, `project-git-credential-attestation-deleted`, `pro…
Vercel added four new values to the `name` field of its `ListEventType` object, which is reachable from 1 Vercel operation: `messageboard-private-project-granted`, `messageboard-private-project-revoked`, `messageboard-pr…
Vercel added a new value, "xaa-updated", to the `name` field on its `ListEventType` object, which is reachable from one Vercel operation, according to the company's changelog. Vercel called the change additive, but it br…
Vercel added three new values to the `name` field of its `ListEventType` object: `connect-cancel-mcp-event-subscription`, `connect-create-mcp-event-subscription`, and `organization-name-update`. Vercel describes the chan…
A developer outlined a guardrail stack for keeping AI coding agents from degrading a codebase, combining ARCHITECTURE.md and AGENTS.md convention files with dependency-cruiser for enforcing allowed import directions betw…
Steve Yegge shut down Gas Town, admitting that despite spending many thousands of dollars a month on coding agent subscriptions, Gas Town was the only thing he ever built with them. Separately, Databricks reported a 60% …
Nearly 1 in 10 publicly accessible LiteLLM instances accept a default master key or require no authentication, exposing cloud AI infrastructure to root-level remote code execution and IAM theft, according to research pre…
Google Threat Intelligence Group (GTIG) reported that in Q2 2026, threat actors compromised a cloud resource and executed an agent-enabled mass credential harvesting campaign in under six hours, highlighting a shift from…