strip_markdown_for_tts (tools/tts_text_normalize.py)

Deterministically strips Markdown formatting for clear TTS output.

# Source: https://github.com/NousResearch/hermes-agent | License: MIT | Copyright 2025 Nous Research
# Provided via justclaw.it.com; keep this notice with the code.
"""Deterministic cleanup turning assistant Markdown into a spoken script.

Shared by explicit TTS calls, gateway auto-TTS, voice-mode streaming and the web
dashboard. Non-ASCII characters are written as escapes on purpose so the file
stays free of invisible/look-alike glyphs.
"""

from __future__ import annotations

import html
import re

# Sentinel appended to former heading lines so smooth_whitespace_for_tts folds the
# heading into the sentence after it ("Weather, it will be sunny") instead of a bare
# "Weather." label.
_HEAD = "\x00"

_MD_CODE_BLOCK_RE = re.compile(r"```[\s\S]*?```")
_MD_LINK_RE = re.compile(r"\[([^\]]+)\]\

... (truncated -- full source via MCP)