Add Discord-native hybrid front-end for Jarvis (bot + bridge)
Some checks failed
Release / semantic-release (push) Successful in 59s
tests / Unit tests (Linux, Python 3.11) (push) Successful in 13m45s
Release / build-linux (push) Failing after 7m47s
Release / build-windows (push) Has been cancelled
Release / build-macos (arm64, macos-latest) (push) Has been cancelled
Release / build-macos (x64, macos-15-intel) (push) Has been cancelled
Release / release-main (push) Has been cancelled
Release / release-develop (push) Has been cancelled

Transform isair/jarvis into a Discord-controlled voice assistant running on
the Ubuntu VNC desktop, keeping the mature ~39k-line Python brain intact.

- bot/ (Node + bun, discord.js): /자비스 slash commands (ephemeral),
  voice channel join + voice receive/playback, pluggable VNC screen broadcast
  (selfbot live / noVNC / screenshot)
- bridge/ (Python, Flask): wraps jarvis STT + run_reply_engine + Piper TTS
  behind a thin localhost HTTP API
- .env.example, scripts/ (start_bridge/start_bot/dev), README rewrite,
  docs/language-comparison.md and docs/vnc-xfce-setup.md

Language decision: hybrid (Python brain + Node/bun Discord layer) because
Discord blocks bot video; native screen broadcast only works via a Node
selfbot library.
This commit is contained in:
javis-bot
2026-06-09 14:51:05 +09:00
parent a5bf8d1826
commit c4abf63f38
308 changed files with 94135 additions and 1 deletions

View File

@@ -0,0 +1,55 @@
from __future__ import annotations
import re
# Deterministic structural scrub patterns. Order matters: specific
# vendor-shaped tokens are matched before generic catches so the more
# informative label wins (e.g. "[REDACTED_AWS_KEY]" beats "[REDACTED_HEX]").
_REDACTION_RULES: list[tuple[re.Pattern[str], str]] = [
(re.compile(r"[A-Za-z0-9._%+\-]+@[A-Za-z0-9.\-]+\.[A-Za-z]{2,}", re.IGNORECASE), "[REDACTED_EMAIL]"),
(re.compile(r"\b(?:\d[ -]*?){13,19}\b"), "[REDACTED_CARD]"),
# Vendor-specific access keys (bare, no surrounding keyword required).
(re.compile(r"\b(?:AKIA|ASIA)[0-9A-Z]{16}\b"), "[REDACTED_AWS_KEY]"),
(re.compile(r"\b(?:sk|pk|rk)_(?:live|test)_[A-Za-z0-9]{16,}\b"), "[REDACTED_STRIPE_KEY]"),
(re.compile(r"\bgh[pousr]_[A-Za-z0-9]{36,}\b"), "[REDACTED_GH_TOKEN]"),
(re.compile(r"\bsk-[A-Za-z0-9]{32,}\b"), "[REDACTED_OPENAI_KEY]"),
(re.compile(r"\bAIza[0-9A-Za-z_\-]{35}\b"), "[REDACTED_GOOG_KEY]"),
# Authorisation headers — Bearer/Basic carry credentials in line.
(re.compile(r"Authorization:\s*Bearer\s+\S+", re.IGNORECASE), "Authorization: Bearer [REDACTED]"),
(re.compile(r"Authorization:\s*Basic\s+[A-Za-z0-9+/=]+", re.IGNORECASE), "Authorization: Basic [REDACTED]"),
# Generic prefix catch — left after the vendor-specific rules so
# newer formats like gh[pousr]_ get a precise label first.
(re.compile(r"\b(AWS|GH|GCP|AZURE|xox[abpcr]-)[A-Za-z0-9_\-]{10,}\b", re.IGNORECASE), "[REDACTED_TOKEN]"),
(re.compile(r"\b(?:eyJ[0-9A-Za-z._\-]+)\b"), "[REDACTED_JWT]"),
# Keyword-anchored credentials. Covers refresh/access/oauth/session
# variants in addition to the original pass/secret/token/apikey set.
(re.compile(
r"\b(pass(?:word)?|secret|token|apikey|api_key|"
r"(?:refresh|access|id|oauth)_?token|session(?:_?id)?|sid)"
r"\s*[:=]\s*\S+\b",
re.IGNORECASE,
), r"\1=[REDACTED]"),
(re.compile(r"\b[0-9A-Fa-f]{32,}\b"), "[REDACTED_HEX]"),
(re.compile(r"\b\d{6}\b(?=.*(otp|2fa|code))", re.IGNORECASE), "[REDACTED_OTP]"),
]
def redact(text: str, max_len: int = 8000) -> str:
scrubbed = text
for pattern, repl in _REDACTION_RULES:
scrubbed = pattern.sub(repl, scrubbed)
scrubbed = " ".join(scrubbed.split())
if len(scrubbed) > max_len:
scrubbed = scrubbed[:max_len]
return scrubbed
def scrub_secrets(text: str) -> str:
"""Apply the structural scrub rules without whitespace collapse or length cap.
Use for structured content (tool output, multi-line payloads) where
preserving newlines matters but tokens/emails/etc. must still be masked.
"""
scrubbed = text
for pattern, repl in _REDACTION_RULES:
scrubbed = pattern.sub(repl, scrubbed)
return scrubbed