
Paid human listeners judge TTS audio blind: pairwise preference or human-vs-AI detection.
Public tool metadata for what this MCP can expose to an agent.
woodstitch_templatesWhat is measured and what is not (no key needed)What is measured and what is not (no key needed)
No parameter schema in public metadata yet.
woodstitch_pricingRate card and worked examples (no key needed)Rate card and worked examples (no key needed)
No parameter schema in public metadata yet.
woodstitch_accountBalance and spendBalance and spend
No parameter schema in public metadata yet.
woodstitch_upload_filesUpload local audio, get file idsUpload local audio, get file ids
No parameter schema in public metadata yet.
woodstitch_create_evalCreate and price a study; dry_run to price onlyCreate and price a study; dry_run to price only
No parameter schema in public metadata yet.
woodstitch_start_evalStart the studyStart the study
No parameter schema in public metadata yet.
woodstitch_get_evalProgress and provisional scoreProgress and provisional score
No parameter schema in public metadata yet.
woodstitch_get_resultsFinal numbersFinal numbers
No parameter schema in public metadata yet.
woodstitch_exportEvery judgement as rows (json or csv, optionally to a file)Every judgement as rows (json or csv, optionally to a file)
No parameter schema in public metadata yet.
woodstitch_cancel_evalStop early, pay only for work doneStop early, pay only for work done
No parameter schema in public metadata yet.
woodstitch_list_evalshousekeepinghousekeeping
No parameter schema in public metadata yet.
woodstitch_list_fileshousekeepinghousekeeping
No parameter schema in public metadata yet.
woodstitch_delete_filehousekeepinghousekeeping
No parameter schema in public metadata yet.
Sponsors
WOODSTITCH_API_KEY*secretWoodstitch API key (ws_live_...), from https://woodstitch.com/app/api-keys. Not needed for the templates and pricing tools.
An MCP server for the Woodstitch API: paid human listeners judge your text-to-speech audio in blinded studies, and your agent gets a number back with a confidence interval.
Two tasks, and only two: pairwise preference (a listener hears two clips of the same text and picks one, forced choice) and human-vs-AI detection (one clip, human or AI, confidence 1–4). Not offered: MOS, CMOS, MUSHRA or any 1–5 scale.
Get an API key at https://woodstitch.com/app/api-keys, then:
Claude Code
claude mcp add woodstitch -e WOODSTITCH_API_KEY=ws_live_... -- uvx woodstitch-mcp
Claude Desktop / Cursor / any MCP client (mcpServers entry):
{
"woodstitch": {
"command": "uvx",
"args": ["woodstitch-mcp"],
"env": {"WOODSTITCH_API_KEY": "ws_live_..."}
}
}
pipx run woodstitch-mcp or pip install woodstitch-mcp && woodstitch-mcp work too.
Requires Python 3.10+.
woodstitch-mcp