6 projects
mizan
The reliability scale for AI agents: restore, balance, classify, constrain, verify, weigh.
toolproof-receipt
Agent tool verification - catch hallucinated tool calls with signed execution receipts, now with MTG (Morphological Type Guards) integration
mtg-guards
Morphological Type Guards — a JSON Schema extension for multilingual tool-call arguments
qadiya
Classify-then-Dispatch for AI agents under constraint-driven canonical cases
muqabalah
Reversible cancellation and normalization with audit trails; fail-loud on contradiction
jabr
Reversible prompt-context restoration with audit trails
