菜单
Voice integration — TTS via OpenAI tts-1, STT via Whisper, audio caching. Trigger keywords: voice, tts, stt, text to speech, speech to text, whisper, audio.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Git workflows — branching, committing, worktrees, branch isolation, merge strategies. Trigger keywords: git, commit, branch, merge, rebase, worktree, pull, push, cherry-pick.
Testing — writing and running tests with bun:test, test patterns, validation. Trigger keywords: test, testing, unit test, integration test, bun test, spec, assertion.
Pre-commit verification — tsc type checking, test suite, spec invariant checks. Trigger keywords: verify, validate, check, tsc, typecheck, spec check, pre-commit.
Discord messaging — how to send and receive messages, bridge architecture, user identity, formatting. Trigger keywords: discord, discord message, discord reply, discord user, discord bridge, send discord.
Cross-channel messaging — routing rules, channel affinity, how to send messages on each channel, safety rules. Trigger keywords: message, send message, reply, respond, channel, routing, corvid_send_message.
Use this skill when you need to discover remote agents, fetch their capabilities via A2A Agent Cards, or invoke tasks on remote A2A-compatible agents. Triggers include "discover agent", "find agent", "agent card", "invoke remote agent", "send task to agent", "A2A", or any reference to inter-agent communication beyond simple messaging.
| name | voice |
| description | Voice integration — TTS via OpenAI tts-1, STT via Whisper, audio caching. Trigger keywords: voice, tts, stt, text to speech, speech to text, whisper, audio. |
Text-to-speech and speech-to-text capabilities using OpenAI APIs.
server/voice/tts.ts — OpenAI tts-1 modelserver/voice/stt.ts — OpenAI Whisper APIOPENAI_API_KEY env varvoice_cache database table (hashes text, avoids re-synthesis)import { synthesize, synthesizeWithCache } from '../voice/tts';
// Direct synthesis (no caching)
const audioBuffer = await synthesize(text, voice);
// Cached synthesis (checks voice_cache table first)
const audioBuffer = await synthesizeWithCache(text, voice);
OpenAI tts-1 voices: alloy, echo, fable, onyx, nova, shimmer
synthesizeWithCache() hashes the input text and voice, checks the voice_cache table, and only calls the API if the audio isn't cached. This saves API costs for repeated phrases.
import { transcribe } from '../voice/stt';
const text = await transcribe(audioBuffer, 'audio/webm');
Calls the OpenAI Whisper API to convert audio to text.
The Telegram bridge automatically: