Blog
Notes and analysis on AI development.
WebMCP turns browser agents from clickers into tool callers
Chrome’s WebMCP proposal lets web pages expose structured tools to browser agents instead of forcing them to infer and click UI controls.
OpenAI adopts SynthID as AI image trust gets a new baseline
OpenAI and Google are turning C2PA, SynthID, and verification tools from image-generator features into web-scale trust infrastructure.
A 75% Discount Becomes the Baseline for DeepSeek V4-Pro
DeepSeek is turning V4-Pro API discount pricing into the new baseline, forcing agent builders to recalculate inference cost and routing strategy.
A 73% Agent Report Card Ends the Model-Only Benchmark Era
Open Agent Leaderboard evaluates full agent systems, not just standalone models, combining architecture, tools, cost, and failure behavior.
agentmemory 0.9.21 turns coding-agent memory into shared infrastructure
agentmemory points at a new layer for coding agents: shared local memory across Claude Code, Codex, Cursor, OpenCode, and other tools.
Claude Gets a Conscience Tool, and Alignment Moves Into the Agent Loop
Anthropic is widening the moral formation conversation around Claude while testing an ethical reminder tool inside the model runtime loop.
Review comments become run buttons in Copilot Agent Merge
GitHub Copilot app, low-cost model routing, Fix batch, and Agent Merge are turning PR review into an agent execution loop.
Goal Mode by default, Codex targets agent waiting time
OpenAI Codex Goal Mode and locked computer use move the coding-agent bottleneck from prompts toward goals, context, approvals, and policy.
Agents Need PCs Too, and Windows 365 Defines the Isolation Layer
Windows 365 for Agents isolates AI agent execution inside Cloud PCs and pairs with Agent 365 governance.
Antigravity 2.0, Google Turns the IDE Into an Agent Command Center
Google Antigravity 2.0 connects the IDE, CLI, SDK, Gemini API managed agents, and Android app generation into one agent operating surface.
Agents compiled into ABL, and the runtime bottleneck Kore.ai is targeting
Kore.ai Artemis frames enterprise agents as compiled ABL artifacts controlled by runtime governance, not loose prompt bundles.
OpenAI’s unit-distance proof puts AI research automation on the record
OpenAI’s counterexample to the Erdős unit-distance conjecture shows both the promise of AI research automation and the reproducibility gap left by an unnamed model.