GTPA: Agency on a Noisy Base Motion, Without Physics. Core Edition

GTPA v2.0 is a mathematical framework for modelling an agent that acts on a noisy base motion. The agent reads the base through a lens, estimates an error, chooses an action by one control law with one effort metric, and adds that action to the base with a dial; at dial zero the base is recovered exactly, and the efficacy of an action is measured against that copy without action. Each statement carries its hypotheses, a status and a check level; worked models are declared as models, and their results hold for those models only. It supersedes GTPA v1.2 (10.5281/zenodo.17994303). v1.2 was audited statement by statement: of 34 verdicts on it, 16 find a statement ill-posed as written, 15 false with a counterexample and 3 true under hypotheses stated in the answer. v2.0 keeps what can be stated exactly and drops every physical reading. The audit, with a disposition for each of the 71 numbered statements of v1.2, is the errata annex (CROSSWALK-v1.2.pdf). The base motion is that of UFD v2.0 (10.5281/zenodo.23171454). The human sector is deferred to v2.1. The proofs cited by number are in the three accompanying files BLOCK-B-ANSWERS.pdf, BLOCK-C-THEOREMS.pdf and BLOCK-C-OBJECTS.pdf. Checks: the text and its declarations were read by independent AI contexts from the same provider as the drafting model (Claude, by Anthropic); the decisive calculations were recomputed by machine, with no disagreement, and the few items with no machine computation rest on those reads; nothing is checked in Lean. It is not physics and it says nothing about any person.