“What did I know and do?”
Retrieves conversations, documents, activity and prior context. Essential for recall — but recall alone does not say which principle should govern a decision.
You already have an AI. Give it the principles, precedents and limits that shape how you make decisions — without training a new model and without surrendering the final call.
TRU does not promise perfect decisions. It makes assumptions visible, alternatives explicit and learning possible.
No account and no card. A free AI reads the public TRU doctrine, proposes a judgment, and returns it only if our decision-contract verifier accepts it. Public questions only; the free lane can occasionally be slow or unavailable.
Ask in your own words. TRU separates facts, doctrine, inference and unknowns; names assumptions, kill indicators and the road not taken; then leaves the decision to you.
From the written record — this reflects what was written and ratified, not a live opinion.
Sources — the ledger behind this answer
TRU would rather expose a gap than invent a position in the owner's name. In the private connector, the question can still be analyzed — but the result must label doctrine, evidence, inference and uncertainty separately.
TRU Mode is not another chatbot and not a permanent personality. It is an owner-invoked decision mode that works inside the AI tools people already use.
Retrieves conversations, documents, activity and prior context. Essential for recall — but recall alone does not say which principle should govern a decision.
Applies ratified principles, confidence thresholds, precedents and limits. It names uncertainty and preserves the road not taken.
Principles, playbooks, decisions and hard-earned experience enter the Foundry as material — never silently as doctrine.
Accept, reject, amend or revoke. Every doctrine pack is versioned and bound to a stable cryptographic reference.
ChatGPT, Claude, Cowork, Codex or another MCP client receives the right doctrine before reasoning. TRU advises; the human decides.
The output is shaped so a human can challenge it now and reality can falsify it later.
The first commercial pilot turns real operating judgment into a private mode inside the AI tools a person or team already uses.
A public doctrine baseline for disciplined, falsifiable decisions.
Your principles and experience, privately formed and owner-ratified.
A shared doctrine layer with roles, ratification and deployment controls.
The address below is a machine endpoint, not a web page. Add it as a custom MCP connector, authorize it, then invoke TRU only when you want it.
https://verify.cotrugli.tech/mcp.tru_mode tool is visible.TRU cannot infer or expand the owner's authority. Serious and irreversible decisions require the human hand.
Cases remain untrusted inputs. Experience becomes doctrine only through a separate, explicit owner-ratified passage.
TRU does not sign, transact, hold market keys or stop the market. It remains advisory even when it fails.
Pack ref: sha256:ff5f659390ead9cf7420fe239dd764f1590ebc54cf0a32638a3f894093bc84b9
Ratified source: sha256:c5f71f5c83a6a0cb59d70adf1c575876c5c8cda597c56d28d62fc851c8ab5f05
Bring ten pages and three real decisions. We will form a private TRU Mode and compare the same cases with and without your doctrine.