https://mcp.getfrenchie.dev/mcp ↗
Frenchie is an MCP-first multimodal utility for AI agents. It helps agents work with files through a hosted MCP server: - OCR PDFs and images into clean Markdown - Transcribe audio and video into Markdown - Generate images from text prompts - More multimodal capabilities are coming soon Frenchie is built for agent workflows, not as a general OCR or speech-to-text platform. Files are processed for the requested job and results expire after delivery. Use it when your agent needs reliable file extraction or generation without building a custom pipeline.
Frenchie is a remote MCP server published at mcp.getfrenchie.dev. It has been probed 7 times since 9/10/2026. It answered in 7 of them (100.0%), a near-uninterrupted record. Median response time is 627 ms, a delay an agent will notice. It offers a narrow, focused set of 7 tools. On the protocol side it still runs 2025-11-25 and has not moved to the newer spec.
Can an LLM agent pick the right tool here — names, descriptions and parameter clarity are assessed.
generate_image — Incomplete parameter explanation (truncated description)upload_file — Ambiguous 'object_key' parameter usageextract_to_markdown — Vague 'structured documents' categorizationtranscribe_to_markdown — Missing explicit error handling for invalid ISO codesget_job_result — No mention of timeout or failure statesRisk: medium
tools/list structure, inputSchema validity, and a functional smoke test — the components of the 0-100 score.
Derived by comparing consecutive probes — changes in era, protocol version, build and reachability.
Add this badge to your README — it updates automatically as measurements change.
[](https://mcpmetrics.io/servers/lab94-frenchie)<a href="https://mcpmetrics.io/servers/lab94-frenchie"><img src="https://mcpmetrics.io/badge/lab94/frenchie/era.svg" alt="mcpmetrics"></a>You are seeing the last 7 days. Sign up for the full history. Which check failed and why is in the dashboard.
Sign up free to see| Run | Era | Modern | ms | Legacy | ms | Versions |
|---|---|---|---|---|---|---|
| 2026-09-11 13:13:33 | Legacy | 400 | 680 | 200 | 678 | 2025-11-25 |
| 2026-09-11 11:20:19 | Legacy | 400 | 679 | 200 | 675 | 2025-11-25 |
| 2026-09-11 09:28:19 | Legacy | 400 | 624 | 200 | 627 | 2025-11-25 |
| 2026-09-11 07:35:54 | Legacy | 400 | 664 | 200 | 665 | 2025-11-25 |
| 2026-09-11 05:44:28 | Legacy | 400 | 613 | 200 | 609 | 2025-11-25 |
| 2026-09-11 03:50:39 | Legacy | 400 | 609 | 200 | 604 | 2025-11-25 |
| 2026-09-10 21:16:35 | Auth-gated | 401 | 325 | 401 | 331 | — |
Each block is one measurement round. Green: working response. Amber: responded but the server was returning errors (5xx). Red: no response at all.
Each cell is one probe run. Faded cells are incomplete probes — one leg did not answer, so the era is inconclusive.
The two probe legs separately: modern server/discover and legacy initialize.