gaelic-ghost/speakswiftlyserver/speak-swiftly-mcp
Claude Code and agent skill by gaelic-ghost/speakswiftlyserver
- community
Checking that the SKILL.md file is still in the repository.
Use when a user wants general help with the slim Speak Swiftly MCP surface, including broad requests to inspect runtime state, read replies aloud, draft voice or text-profile guidance, or decide whether to use MCP resources, MCP prompts, the live-speech MCP tool, or HTTP endpoints.
Install
npx vskill@latest install gaelic-ghost/speakswiftlyserver/speak-swiftly-mcpSKILL.md
SpeakSwiftly MCP
Use this skill as the first orientation pass when the request is about the SpeakSwiftly MCP surface broadly rather than one narrow operation.
Start Here
- Treat the local
speak_swiftlyMCP server from this repository's `.mcp.json` as the default surface. - For a broad status question, read
speak-swiftly://overviewfirst. Useget_runtime_overviewonly for compatibility clients that cannot read MCP resources cleanly. - For a "what can this surface do?" question, use API.md first, then the current source-of-truth catalog files:
- MCPToolCatalog.swift
- MCPResources.swift
- MCPPrompts.swift
Workflow Split
- Runtime, queue, playback, request tracking, backend switching, and cancellation over HTTP plus slim MCP reads:
Use $speak-swiftly-runtime-operator.
- Voice creation over HTTP, voice selection, live speech over MCP, retained artifact generation over HTTP, and retained artifact inspection over HTTP:
Use $speak-swiftly-voice-workflows.
- Text normalization styles, stored text profiles, and replacement authoring over HTTP plus MCP drafting prompts:
Use $speak-swiftly-text-profiles.
General Operating Rules
- Prefer MCP resources for orientation and verification. Use the MCP tool only for immediate live speech playback through
generate_speech; use HTTP endpoints for mutations, cancellation, clearing, playback control, retained generation, runtime changes, and network-audio selection. - When the user needs help deciding which action family fits best, use the
choose_surface_actionprompt instead of improvising from memory. - When a tool returns
request_id, follow it withspeak-swiftly://requests/{request_id}orspeak-swiftly://overviewinstead of guessing whether the work finished. - Distinguish generation backlog from playback backlog. A request can be done generating and still be queued for playback.
- Do not silently substitute a different voice profile when the requested profile is missing unless the user explicitly asks for fallback behavior.
- When repository docs and the live MCP server seem out of sync, trust the current source files in
Sources/SpeakSwiftlyServer/MCP/over older prose.
Security
- Security scanPassed, 95 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingskills/speak-swiftly-mcp/SKILL.md
Popularity
- Repository stars
- 8
- Repository forks
- 2
- 7-day trend score
- 11
Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/gaelic-ghost/speakswiftlyserver/speak-swiftly-mcp)