Pre-built agent toolboxes break the moment a client asks something you did not plan for. Here is the plan-code-execute setup I use instead, where the agent writes its own Python, plus the hallucinated CSV that taught me why the analyst step is mandatory.
Freshly Pressed is back in the WordPress.com Reader, with people picking the posts instead of an engagement score. If you are self-hosted, that only helps once your Jetpack sync is actually working, so here is what to check.
Chat history arrays are a log, not memory. This walks through the LLM memory layer I build instead: DSPy signatures for extracting atomic facts, QDrant for per-user vector storage, and a maintenance agent that decides when an old fact gets updated or deleted.
Fifteen real WordPress sites looked at as architecture rather than inspiration: which themes they run, how they handle search and subscriptions, and where the block editor quietly does the work a plugin usually gets credit for.
The February 4, 2026 dev chat put client-side media on the road to WordPress 7.0: libvips compiled to WASM, running in a worker thread, with image resizing done in the visitor’s browser instead of on your server.
WordPress is testing Gutenberg Content Guidelines, a machine-readable place to keep brand voice, accessibility and image rules so plugins and AI tools can read them instead of guessing from a PDF nobody opens.
The WP AI Client merge proposal would put a provider-agnostic AI API in WordPress 7.0: one PHP interface, a fluent prompt builder, WP_Error returns, and no outbound calls until a plugin configures a provider.
Notes from the WordPress performance chat: bridging classic scripts with script modules for the CodeMirror upgrade, memoizing wp_normalize_path, and why Copilot kept reintroducing a PCRE bug that had already been fixed.
The Core AI team put out a Call for Testing for the AI Experiments plugin. Here is what is in it: MCP server support, markdown feeds, extra providers and AI request logging, plus the $400 API bill that convinced me logging is the part you cannot skip.
A million-token context window does not mean a model can reason across a million tokens. Recursive Language Models hand the context to a sandboxed Python REPL and let the model write code to partition and query it, which is how DSPy now implements the strategy.