What is speech-mcp AGENTS.md?

A repository instruction file for speech-mcp, a speech gateway that connects coding agents and applications to speech, voice-cloning, wake-word, and retrieval services. It describes the project's architecture, providers, ports, environment variables, and coding rules.

Use it when working on the speech-mcp repository. It helps locate the Vite front end, FastAPI back end, MCP SSE endpoint, provider configuration, LanceDB retrieval data, and local wake-word service.

It gives coding agents the project context needed to change the repository without guessing how its front end, back end, MCP connection, speech providers, and database fit together. It also records required language and style conventions.

How sandraschi/speech-mcp configures Codex and OpenCode

Reuse

Borrowing it

Nothing to install: this file belongs to sandraschi/speech-mcp. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/sandraschi/speech-mcp/main/AGENTS.md

Clone the repo

git clone --depth 1 https://github.com/sandraschi/speech-mcp

Made for: Codex, OpenCode.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/instructions/sandraschi/speech-mcp/agents-md/github.svg)](https://agentmods.dev/instructions/sandraschi/speech-mcp/agents-md)

Your own site

<a href="https://agentmods.dev/instructions/sandraschi/speech-mcp/agents-md"><img src="https://agentmods.dev/badge/instructions/sandraschi/speech-mcp/agents-md/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/instructions/sandraschi/speech-mcp/agents-md"><img src="https://agentmods.dev/badge/instructions/sandraschi/speech-mcp/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 1,384 This file is loaded in full into every session.

When invoked 1,384 The same file — it is already loaded in full.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.01384	$0.01384
Opus 5.5	$0.00554	$0.00554
Sonnet 5.5	$0.00277	$0.00277
Haiku 4.5	$0.00138	$0.00138

Measured 1mo ago against content hash b17b466c13b2, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

speech-mcp AGENTS.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

AGENTS.md · 125 lines

How it starts

The opening of the file, as written. The whole thing — 125 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENTS.md — speech-mcp

Rules for AI coding agents (Claude, Cursor, Windsurf, Goose) working on this repo.

Project Identity

Name: speech-mcp
Purpose: Multi-provider speech gateway — Gemini Live, Hume AI, ElevenLabs TTS + voice cloning
Owner: Sandra Schipal, Vienna
Fleet role: Voice AI bridge for the entire fleet — TTS, STT, wake word, RAG

Architecture Quick Reference

FastMCP stdio server  <──>  Claude Desktop / MCP clients
       │
FastAPI REST :10909  <──>  React/Vite frontend :10908
       │
MCP SSE :10909/mcp  <──>  Bridge target for ProxyProvider
       │
Providers (configured via .env)
  ├── Gemini 3.1 Flash TTS      (GOOGLE_API_KEY)
  ├── Hume AI Octave + EVI      (HUME_API_KEY)
  ├── ElevenLabs TTS + Cloning  (ELEVENLABS_API_KEY)
  ├── Gemma 4 local             (no key required)
  └── Windows SAPI5             (no key required)
       │
LanceDB (RAG)  ←  data/lancedb/
       │
openWakeWord (local wake word)  ←  no API key required

Ports

Service	Port
Frontend (Vite)	10908
Backend (FastAPI + MCP SSE)	10909

Code Rules

Python

Python 3.12+ — use modern type hints (str | None, TypeAlias)
Async-first: all MCP tools must be async def — no blocking on event loop
Type annotations: all tool params use Annotated[T, Field(description="...")] — no Args: blocks
No global state except _timers and _log_queue in server.py
Logging: use logging.getLogger(__name__) — never print()
Line length: 120 chars (ruff enforces)

Tool Registration

Tools registered via @mcp.tool() decorator in domain modules under src/speech_mcp/tools/
Portmanteau imports re-exported in src/speech_mcp/tools/__init__.py
All tools return {"success": bool, ...} dict
List/status/stats tools use @mcp.tool(app=True) with Prefab UI
Long-running tools use @mcp.tool(task=True) (background tasks)

Docstrings

Summary: 1-3 lines
## Return Format: explicit JSON structure
## Examples: 1-3 concrete calls
No Args: blocks — use Annotated in signatures

Read the full file on GitHub · 125 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

1mo ago First seen · 125 lines · 1,384 tokens per session scan A b17b466c13b2

Subscribe to this mod's changes

speech-mcp AGENTS.md is an instructions file published in the GitHub repository sandraschi/speech-mcp (2 stars, last pushed yesterday), licensed MIT. It adds 1,384 tokens to every session, about $0.0055 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

elevenlabs fastmcp funasr mcp speech-to-text text-to-speech

vscode buildNext.instructions.md

Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).

microsoft/vscode · 9,715 tokens

speech-mcp: Instructions file for Codex