markitdown
[ tools: 1 ][ checks: 13 ][ latency: 5ms ]
73/100(C)
uvx markitdown-mcp
check breakdown
✓Tools exposed — Server exposed 1 tool.
• convert_to_markdown
✓Input schemas valid — All tools have a valid JSON-Schema input definition.
!Input schema quality — 2 schema-quality issue(s) across 1 tool(s).
convert_to_markdown.uri: parameter has no description
convert_to_markdown: inputSchema.additionalProperties is not set to false (extra args pass through silently)
→ Declare `required` for every schema that has properties, so agents know which params are mandatory.
→ Give every property a `description` and a concrete `type` (avoid bare `{}` placeholders).
→ Set `additionalProperties: false` on the top-level inputSchema so models can't silently pass extra, unvalidated arguments.
!Output schemas — No tools declare an outputSchema.
→ Optional: declare outputSchema for tools with structured results so agents can validate/parse output reliably.
✓Duplicate tool names — All tool names are unique.
✓Tool naming — All tool names use spec-legal characters.
!Descriptions — 1 description issue(s) found.
convert_to_markdown.uri: parameter has no description
→ Descriptions are the model's only guide to when/how to call a tool — treat them as prompt engineering.
→ Describe every tool and every parameter, including units, formats, and constraints.
→ Each tool's description should say something the name doesn't, and should be unique to that tool.
!Tool annotations — Annotation coverage 0% (0/5 hint fields set).
• convert_to_markdown: no annotations at all
→ Add annotations: title, readOnlyHint, destructiveHint, idempotentHint, openWorldHint.
→ These let clients gate risky calls and give agents/users a friendly display name — optional but valuable.
!Tool titles — 1 of 1 tool(s) have no title.
• convert_to_markdown: no top-level title (or annotations.title)
→ Declare a top-level `title` on each tool — it gives clients a friendly display name distinct from the machine-readable `name`.
→ `annotations.title` (the older location) is also accepted, but the top-level `title` field is now the spec-preferred place for it.
!Server instructions — No top-level instructions declared.
→ Declare instructions to guide agents on when/how to use this server.
→ Return a non-empty `instructions` string in your InitializeResult — most SDKs let you set this alongside serverInfo.
✓Handshake + list latency — tools/list responded in 5ms.
!Prompts — Server declares the prompts capability but exposes 0 prompts.
→ If you don't actually offer prompts, remove the `prompts` capability from your initialize response.
→ Otherwise, register at least one prompt via your prompts/list handler.
!Resources — Server declares the resources capability but exposes 0 resources.
→ If you don't actually offer resources, remove the `resources` capability from your initialize response.
→ Otherwise, register at least one resource (or resource template) via your resources/list handler.
add the badge
Paste into your README:
[](https://scoremcp.dev/servers/markitdown-mcp)