Concepts
Concepts
How usertrust governs every LLM call — the banking patterns, cryptographic proofs, and policy engine under the hood.
usertrust combines financial accounting patterns with cryptographic verification to make every AI agent call auditable and budget-safe. These pages explain the core mechanisms.
Two-phase spendBanking-style authorization holds for atomic budget enforcementAudit trailSHA-256 hash-chained, append-only log for every governed callMerkle proofsRFC 6962 inclusion and consistency proofs for public verifiabilityLocal modelsHow usertrust classifies local endpoints and meters Ollama, vLLM, and LM Studio at nominal ratesPolicy engine12 field operators, scope globs, and time windows for rule-based governancePattern memorySHA-256 prompt-hash history of model, cost, and outcome; suggestModel() returns the model with the best cost-adjusted success ratio for the caller to route onCircuit breakerPer-provider failure isolation with automatic recovery