Spaces:
Running
Running
Deploy Aureon OS (deploy/selfhost stack: nginx edge + fusion API x2 + console + HNC organ + redis + postgres; paper mode) (part 3)
1ccf652 verified | """Code-writing expression context for Aureon's self-authoring systems. | |
| This module reuses the whole-knowledge voice core for code work. It keeps | |
| generated skills and reviewable patch proposals grounded in repo evidence, | |
| runtime state translation, and readable intent without changing execution | |
| semantics. | |
| """ | |
| from __future__ import annotations | |
| import json | |
| import time | |
| from pathlib import Path | |
| from typing import Any, Dict, Optional | |
| CODE_EXPRESSION_SCHEMA = "aureon_code_expression_context_v1" | |
| def build_code_expression_context( | |
| goal: str, | |
| *, | |
| evidence: Optional[Dict[str, Any]] = None, | |
| profile: Any = None, | |
| root: Optional[str | Path] = None, | |
| evidence_dir: Optional[str | Path] = None, | |
| publish: bool = True, | |
| ) -> Dict[str, Any]: | |
| """Return a human-readable evidence context for a code proposal. | |
| The returned dict is intentionally safe to embed in SkillProposal, | |
| Skill, and SafeCodeControl JSON. It stores compact public summaries, | |
| not raw source text or secrets. | |
| """ | |
| context: Dict[str, Any] = { | |
| "schema_features": [CODE_EXPRESSION_SCHEMA], | |
| "ok": False, | |
| "goal": str(goal or "code proposal"), | |
| "created_at": time.time(), | |
| "voice_summary": "", | |
| "runtime_summary": "", | |
| "top_facets": [], | |
| "source_count": 0, | |
| "novelty_checks": {}, | |
| "redaction_applied": False, | |
| "evidence_path": "", | |
| "warnings": [], | |
| } | |
| try: | |
| from aureon.vault.voice.whole_knowledge_voice import compose_voice_artifact | |
| artifact = compose_voice_artifact( | |
| goal or "code proposal", | |
| audience="developer", | |
| mode="conversation", | |
| evidence=evidence or {}, | |
| profile=profile, | |
| root=root, | |
| evidence_dir=evidence_dir, | |
| publish=publish, | |
| ) | |
| runtime_translation = artifact.runtime_translation or {} | |
| profile_summary = artifact.profile_summary or {} | |
| context.update( | |
| { | |
| "ok": bool(artifact.ok), | |
| "voice_summary": _compact_text(artifact.text, limit=700), | |
| "runtime_summary": _compact_text(str(runtime_translation.get("summary") or ""), limit=360), | |
| "senses": runtime_translation.get("senses", {}), | |
| "blockers": runtime_translation.get("blockers", []), | |
| "top_facets": profile_summary.get("top_facets", []), | |
| "source_count": int(profile_summary.get("source_count") or 0), | |
| "novelty_checks": artifact.novelty_checks, | |
| "redaction_applied": bool(runtime_translation.get("redaction_applied")), | |
| "evidence_path": artifact.evidence_path, | |
| "warnings": list(artifact.warnings or []), | |
| } | |
| ) | |
| except Exception as exc: | |
| context["warnings"].append(f"code_expression_unavailable:{exc}") | |
| if publish: | |
| try: | |
| out_dir = Path(evidence_dir) if evidence_dir else Path(root or ".") / "state" | |
| out_dir.mkdir(parents=True, exist_ok=True) | |
| out = out_dir / "aureon_code_expression_last_run.json" | |
| context["evidence_path"] = str(out) | |
| out.write_text(json.dumps(context, indent=2, sort_keys=True), encoding="utf-8") | |
| except Exception as exc: | |
| context["warnings"].append(f"code_expression_publish_failed:{exc}") | |
| return context | |
| def _compact_text(text: str, *, limit: int) -> str: | |
| clean = " ".join(str(text or "").split()) | |
| if len(clean) <= limit: | |
| return clean | |
| return clean[: max(0, limit - 3)].rstrip() + "..." | |