15
Published Tools
0
Total Stars
4,191
Weekly Downloads
2
GitHub Followers
2
Public Repos
Published Tools
4 MCP Servers11 Skillsacross 5 categoriesMichal Makowski
Anthropic (Claude) LLMProvider plugin for moxxy. Streams the Messages API with tool use + adaptive thinking; translates moxxy ProviderRequest/Event ↔ the Anthropic wire shape. Register it into an @moxxy/core Session to route turns to Claude models.
...moreMichal Makowski
OpenAI (and OpenAI-compatible: z.ai, xAI, Google, Ollama/local) LLMProvider plugin for moxxy. Streams Chat Completions with tool_calls; translates moxxy ProviderRequest/Event ↔ the OpenAI wire shape. Register it into an @moxxy/core Session to route turns
...moreMichal Makowski
MCP (Model Context Protocol) client plugin for moxxy — connect to MCP servers and surface their tools.
Michal Makowski
OpenAI text-to-speech Synthesizer for moxxy. Text → spoken audio bytes via POST /v1/audio/speech. Plugs into the session's SynthesizerRegistry; read-aloud surfaces (desktop 'Read aloud', a voice channel) consume it transparently. The first shipped Synthes
...moreMichal Makowski
Claude subscription (Pro/Max) LLMProvider plugin for moxxy. Talks to the standard Anthropic Messages API using a Claude Code OAuth token from `claude setup-token` or an in-app sign-in.
...moreMichal Makowski
OpenAI Codex (ChatGPT Pro/Plus OAuth) LLMProvider plugin for moxxy. Streams against https://chatgpt.com/backend-api/codex/responses using PKCE-based OAuth tokens.
...morewebeferen
Typed public surface for the moxxy framework: event types, define* factories, lifecycle hook signatures.
webeferen
Google Gemini Flash-Lite text-to-speech extension for moxxy.
webeferen
ElevenLabs text-to-speech Synthesizer for moxxy. Text → spoken audio bytes via POST /v1/text-to-speech/{voiceId}. Plugs into the session's SynthesizerRegistry; read-aloud surfaces (desktop 'Read aloud', a voice channel) consume it transparently. The highe
...morewebeferen
Voice/TTS control for moxxy. set_voice / list_voices let the agent switch which text-to-speech backend read-aloud surfaces use (a registered synthesizer, or 'system' for the OS voice) without a settings UI.
...morewebeferen
Fully local, on-device text-to-speech Synthesizer for moxxy. Runs Piper voices (English + Polish) via sherpa-onnx in a forked sidecar process; voice models download once on first use from sherpa-onnx's pinned releases (sha256-verified). No API key, no net
...morewebeferen
Web capabilities for moxxy: pluggable web_search fallback, web_fetch, and a Playwright browser_session sidecar.
webeferen
Default loop strategy (mode) for moxxy — a Claude Code-style ReAct tool-use loop. Register it into an @moxxy/core Session to give your agent its turn-by-turn reasoning + tool-calling behaviour.
...morewebeferen
A shared terminal surface for moxxy: an embedded PTY the user and the agent drive together (the `terminal` tool runs commands; the desktop renders the live session).
...morewebeferen
Slack bot channel for moxxy. Ingests Slack Events API over the proxy relay, drives the Session, and streams threaded replies. Autonomous allow-list permission model (no human in the loop).
...more