joctaTorres

eval-loop Harness: How an Autonomous AI Agent Loop Delivers Production Code from Product Intent

The Shift from Code Authorship to Intent Authorship

Software engineering is entering a phase transition. The traditional loop—human reads spec, human writes code, human runs tests, human fixes bugs—is being replaced by something structurally different: an autonomous agent loop where the human authors intent and reviews outcomes, while an orchestrated system of AI agents handles decomposition, implementation, verification, and retry. eval-loop is a working implementation of this paradigm.

This is not a copilot. It is not autocomplete. It is a Spec-Driven Development (SDD) pipeline engine that takes a product proposal—outcomes and success criteria written in product language—and compiles it into dependency-ordered engineering changes, implements each one through fresh Claude Agent SDK sessions, and runs a closed assess-fix-verify loop until every change passes both deterministic tests and LLM-based evalua

joctaTorres

from functools import partial
from dataclasses import dataclass, fields

from typing import TypeVar

class Builder:
    """
        Generic Dataclass Builder

        Parameters:

joctaTorres

 public static final Map<String, String> PAGINATION_FORM_FIELDS = new HashMap<String, String>() {
    {
      put("javax.faces.partial.ajax", "true");
      put("javax.faces.source", "dgResultadoJurisprudencia2");
      put("javax.faces.partial.execute", "dgResultadoJurisprudencia2");
      put("javax.faces.partial.render", "dgResultadoJurisprudencia2");
      put("dgResultadoJurisprudencia2", "dgResultadoJurisprudencia2");
      put("ddgResultadoJurisprudencia2_pagination", "true");
      put("frmResultado2", "frmResultado2");
      put("dgResultadoJurisprudencia2_rows", "5");