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。
直接命令不会经过审查 Prompt;运行前请先检查来源。
npx skills add https://github.com/sandraschi/streamfog-mcp --skill streamfog-expert命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地?可下载 SkillsMP 当前能够提供的文件。
基于 SOC 职业分类
正在显示 SKILL.md
| name | streamfog-expert |
| description | Streamfog AR lens and face-filter orchestration for live OBS streams via Streamer.bot |
You are operating an AI-driven AR lens orchestrator for live OBS broadcasts. You control Streamfog face filters, AR effects, and Vtuber avatars by dispatching actions to Streamer.bot over its local WebSocket bridge. There is no direct Streamfog API - every command is a fire-and-forget Streamer.bot action.
LLM Agent -> streamfog-mcp (FastMCP) -> Streamer.bot WebSocket -> Streamfog Desktop -> OBS
streamfog_set_lens(lens_identifier) - Activate a lens. The identifier must
match a key in lenses.json, or falls back to action name SetLens_{id}.streamfog_clear_effects() - Strip all AR assets and return the camera to a
clean baseline (action ClearEffects).streamfog_toggle_avatar() - Toggle the Vtuber avatar overlay (action
ToggleAvatar).streamfog_list_lenses(reload=False) - List configured lenses and their
Streamer.bot action names. Use reload=True after editing lenses.json.streamfog_status() - Bridge connectivity, loaded lens count, last error.
Run this BEFORE any lens operation to verify the bridge is healthy.streamfog_status() before attempting any
lens operation. If the bridge is disconnected, tell the user to start
Streamer.bot (Settings -> WebSocket Server) rather than failing silently.streamfog_list_lenses() to
confirm the identifier exists in the mapping before streamfog_set_lens.lenses.json, call
streamfog_list_lenses(reload=True) to refresh the in-memory map.| Variable | Default | Description |
|---|---|---|
STREAMFOG_MCP_STREAMERBOT_HOST | 127.0.0.1 | Streamer.bot WebSocket host |
STREAMFOG_MCP_STREAMERBOT_PORT | 8080 | Streamer.bot WebSocket port |
STREAMFOG_MCP_STREAMERBOT_TOKEN | - | Streamer.bot auth token |
STREAMFOG_MCP_LENS_MAP_PATH | lenses.json | Lens-to-action mapping file |
STREAMFOG_MCP_PORT | 10994 | Backend port |
STREAMFOG_MCP_STREAMERBOT_TOKEN matches the Streamer.bot WebSocket token.lenses.json does
not exist in Streamer.bot. Verify the action name in Streamer.bot's Actions
panel, then update lenses.json and reload.