Changelog
Every entry below is a real ship date from the build record. This is what the context layer for humans and their agents looks like being built: state, receipts, and no silent rewrites, in the product and in this page.
Every operating constant, timeouts, batch sizes, thresholds, was audited and promoted to administrative configuration with automatic drift checks.
- Tuning changes no longer require a deploy, and the running values cannot quietly diverge from the documented ones.
Tool access is now derived from one canonical registry, scoped to the session, rebuilt from a single source file on every cold start.
- The registry is never hand-edited, and drift between code and registry is detected automatically.
Lock a project brief into a preset code, and any AI session can load the full combination in one call.
- A preset bundles your profile, working rules, memory, mistakes, and one file or topic, with live token counts on every slot.
- Paste the code into Claude, ChatGPT, or your own agent: same brief, every time.
The whole map is now inspectable and editable from a home dashboard, and starring a row guarantees it loads into every session.
- Edit or correct any atom; nothing is ever hard-deleted, so every correction is reversible.
- What you star is what your sessions always know.
Context Windows Close. AI Forgets Everything. Your Work Should Never Start From Zero.
MrAgentˣ is in private beta. Limited to the first 1,000 until launch.
Join the waitlist