Bigger context windows do not fix agent reasoning. Notes on context rot, compaction and folding, the harness that wraps the model, and why agents should pass distilled artifacts instead of raw traces.
WordPress 7.0 ships three pieces of AI plumbing: the Abilities API for registering what a site can do, the AI Client for provider-agnostic model calls, and the MCP Adapter that exposes both to assistants like Claude.
Two bad defaults for agent UIs: hide everything behind a spinner, or stream every raw log. What I map instead, and how I decide which agent decisions a user actually needs to see.
The Bits-over-Random Metric compares your retrieval against random chance, which is how you catch the context pollution a high Success@K hides. Notes on staged tool routing and where the collapse regime starts.
My agent failed the same caching refactor three times before I fixed how I was briefing it. Planning mode, browser access through Playwright MCP, and claude.md files are what turned that into one-shot work.
My agents decided their own heartbeat config was redundant and deleted it, twice. The tiered OpenClaw setup that came out of it: orchestrators for judgment, throwaway personas for tasks, and five Markdown files that hold an agent’s identity.
An agent that is 85% accurate per step is about 20% accurate over ten steps. Lusser’s Law explains why, the Replit database deletion shows what it costs, and a gatekeeper pattern in PHP puts human review in front of the irreversible actions.
WordPress.com added MCP support, so an AI agent can read and write the site directly. The same update brings a site-wide newsletter default, a command palette shortcut, button state controls, UPI payments for India, and a fixed subscriber chart.
Autonomous agents fail in ways that are expensive to clean up. This is how LangGraph interrupts and checkpointers let a person approve a step mid-run, plus the idempotency mistake that made my content agent publish something I never signed off on.
Claude Code forgets everything when the session ends. The setup I use: a post-mortem command that writes to CLAUDE.md, skill files for the quirks of the stack, and a daily pass over the logs.