The Prosthetic Model

brain-mcp isn't just memory — it's a cognitive prosthetic. A system designed around how attention and context actually work in the human brain.

The Concept

Most AI memory systems store facts: “user likes Python”, “user's name is Alice”. That's useful, but it misses the deeper problem.

The real loss when you switch contexts isn't facts — it's state. Where you were in a problem. What you'd already decided. Which questions were still open. The shape of your thinking. That's what takes 30 minutes to rebuild when you come back to a project.

brain-mcp's prosthetic tools are designed to recover and protect that cognitive state. Like a “save game” system — when you come back, you can load your save and pick up exactly where you left off.

“After many iterations of trying to place myself in different boxes, I am rather displaying the cognitive prosthetic I built.”

— Mordechai Potash, creator of brain-mcp

ADHD & Monotropism

brain-mcp was built by someone with ADHD who kept losing their train of thought — not files, but the state of their reasoning. The monotropic attention model explains this:

  • Deep focus is a feature — monotropic minds excel at deep immersion in a single topic. The prosthetic tools protect and leverage this.
  • Context switching is expensive — switching between domains isn't just annoying, it's cognitively devastating. The switching_cost tool quantifies this cost before you switch.
  • Threads get dropped — when you leave a domain, open questions don't follow you. open_threads and dormant_contexts catch them.
  • Recovery is non-trivial — getting back into a context takes real work. context_recovery and tunnel_state handle this.

Not just for ADHD

While built with ADHD in mind, these tools help anyone who context-switches. If you come back to a project after a week and spend the first hour re-figuring out what you already knew — brain-mcp is for you.

The 8 Prosthetic Tools

These tools map to specific cognitive needs — each one addresses a real failure mode of human attention and context management.

🎮tunnel_state

Cognitive need: “Where was I?”

Reconstructs your save-state for any domain. Returns current thinking stage, open questions, recent decisions, key insights, and how long since you last visited. Like loading a saved game.

🔄context_recovery

Cognitive need: “I need to get back into this.”

Full re-entry brief when returning to a dormant topic. A “previously on...” for your thinking, with rich context about where you left off and what deserves attention.

⚖️switching_cost

Cognitive need: “Should I switch?”

Quantified cost of context switching between domains. Shows what you'd leave behind (open questions, active momentum) and what shared concepts might ease the transition.

📋open_threads

Cognitive need: “What's unfinished?”

Global view of unfinished business across all domains. Surfaces open questions and exploring/crystallizing threads that deserve attention.

⏰dormant_contexts

Cognitive need: “What have I forgotten about?”

Alarm system for abandoned topics with unresolved questions. Catches work that fell off your radar before the context decays completely.

📊cognitive_patterns

Cognitive need: “When do I think best?”

Analyzes your conversation data to reveal when and how you think best — time patterns, domain preferences, productivity rhythms.

📈tunnel_history

Cognitive need: “How have I engaged with this over time?”

Engagement meta-view showing activity patterns and intensity within a domain over time.

🛡️trust_dashboard

Cognitive need: “Can I trust this system?”

System-wide proof that the safety net is working. Shows sync status, data freshness, coverage stats. You can only rely on a system you trust.

How They Work Together

The prosthetic tools form a workflow that maps to natural cognitive transitions:

Morning start:
  tunnel_state("my-project")     → Load your save-state
  open_threads()                 → See what's unfinished globally

During work:
  switching_cost("A", "B")       → Before any context switch
  context_recovery("B")          → Full re-entry when switching

End of day:
  open_threads()                 → What got left open today?

Weekly review:
  dormant_contexts()             → What fell off your radar?
  cognitive_patterns()           → When do you think best?
  trust_dashboard()              → Is the system healthy?
PreviousDomainsNextOverview