All activity
Tommaso Giacomo Bredariolleft a comment
We’ve been pretending LLMs “think” for too long....This project does something uncomfortable: it strips that illusion away and turns reasoning into something you can actually control, constrain, and break on purpose. Just a substrate executing instructions—and an architecture that decides what “thinking” is allowed to look like. People will either hate this… or build their entire stack on it😅🥀
Meta-ReasoningYour LLM doesn't think. We make sure of it.
Meta-Reasoning is an SDK for governing how LLMs reason. Instead of treating models as agents, it externalizes cognition into a controllable architecture: observe reasoning structure, mutate it with constraints, and record trajectories in an epistemic ledger. Enable debugging, replay, benchmarking, and CI/CD for reasoning. Make LLM behavior testable, deterministic, and governed.
Already available on 🦞OpenClaw, 🟠ClaudeCode and 🤖Codex!
Meta-ReasoningYour LLM doesn't think. We make sure of it.