Held Out Structural Decoding of Four Solana Program Interfaces

This report analyzes an archived evaluation of four interface artifacts returned by a Composer service for Solana programs without an IDL found in the checked publication channels. Each artifact was frozen and hashed before collecting later transaction signatures. The verifier tests selector, account count, payload length, fixed account identity, applicable account flags, program derived address recipes, and byte round trips while leaving argument semantics opaque. The historical records contain 125 invoking transactions and 198 tested instructions: 84 structurally match and 114 do not. Per program coverage ranges from zero of 32 to 29 of 30. Seventy one program derived addresses are reported as verified in the one successful PDA bearing family. The result is heterogeneous structural coverage, not semantic decompilation or proof of the learner's algorithm. This manuscript inspects existing artifacts and verifier source without rerunning the experiment. It preserves failures, distinguishes missing catalogs from unavailable services, and separates signed transaction authenticity from provider supplied execution metadata. The evaluation supports transparent evidence practices for tooling and accountability, while claims about user empowerment, financial access, or detection of wrongdoing require additional direct evidence. Preparation of this report was assisted by Codex. The research programme is directed toward human agency, financial inclusion, and evidence-based accountability. Intended social benefits are distinguished from measured technical results in the manuscript. The manuscript and newly authored analysis are offered under CC BY 4.0. Any supplied source files retain the license terms stated in the supplement; the record license does not replace third-party source licenses.