AI Assist
Freemium
Users prioritizing on-device processing and data privacy
Every time you switch LLMs, you start over — repasting context, re-explaining your projects, re-establishing your voice from scratch. myPKA solves that by making the folder, not the model, the source of truth.
The system is a local markdown folder pre-loaded with a six-person AI team: a routing orchestrator (Larry), a research specialist (Pax), a capture agent (Penn), and others — each with a named contract and a session journal so the next model picks up where the last one left off. You bring your own LLM; the folder supplies the memory. Research produces structured notes in place, drafts inherit your established voice, and weekly review prompts surface stale items automatically. The ceiling appears when you need real-time data, API integrations, or collaborative editing — none of that is in the folder. Teams that need those reach for purpose-built tools alongside this one.
Bottom line: Pick this if you switch LLMs often and need context that survives the switch; plan a different architecture if your workflows require live API calls, multi-user collaboration, or anything that can't live in a markdown file.
myPKA folder with six specialists, open source under CC BY-NC-SA 4.0
Course, coaching, community, and expansion packs
View full pricing on myicor.com →
Pricing may have changed since last verified. Check the official site for current plans.
Find your fit
More options in Productivity. Compare pricing, platforms, and features side by side.
Freemium
Users prioritizing on-device processing and data privacy
Free
Users seeking culturally grounded AI responses
Most AI productivity setups collapse the moment you change models — every new session is a blank slate that knows nothing about your projects, your voice, or your team’s history. myPKA is a local folder of plain markdown files that carries that context forward regardless of which LLM you open it with. Inside the folder is a pre-built team scaffold: Larry routes every request to the right specialist, Pax runs structured research, Penn handles capture, Nolan manages the hiring procedure when you want to add new agents, and Mack and Silas cover connections and database architecture. The vendor describes the model as ‘the keyboard’ and the folder as ‘the database’ — you swap keyboards; the database stays put.
The key differentiator is LLM-agnosticism enforced at the file level. Because every agent contract, journal, and project note is a plain text file on your disk, Claude, Gemini, Codex, Cursor, or any future model can read it without migration, without a cloud account, and without vendor lock-in. The vendor licenses the base scaffold under CC BY-NC-SA 4.0, which means the folder structure and agent contracts are open source — your data remains yours even if the company stops operating. MCP integration is available for members, which the vendor describes as removing the need to manually keep configuration files updated.
This fits solo operators and small teams who want an always-available AI staff that accumulates institutional memory across sessions and model switches. It breaks down when your work requires live data fetches, webhooks, multi-user simultaneous editing, or workflows that span systems outside the folder. Teams doing real-time integrations or needing a shared cloud workspace will hit that wall quickly. The folder-plus-agent model also requires you to supply your own LLM access — there is no hosted inference or managed API; you wire up the model yourself.
The paid membership gates the ICOR methodology course, coaching, community access, and expansion packs like the Designer Pack. The base folder scaffold — the agents, contracts, and folder structure — is free and open source. Self-hosting in the literal sense is inherent: the folder runs on your local disk with no server required.
Sign in to rate this tool or suggest an edit.