Answers from the documentation; checking which model is on duty
Hi - I answer from the OpenSmartRoute documentation: routing, the API, plans and quotas, self-hosting. Ask away, or open a support ticket if you need a person.
Grounded in the docs - follow a source before acting on it.
docling - Skill - OpenSmartRoute
Skillv1.0.0
docling
Convert PDFs, Office documents, HTML pages, Markdown files, images, and scanned documents into structured Markdown or JSON.
Imported from Rycen7822/skill-retrieval-mcp (tests/fixtures/retrieval_quality/skills/user-imported/docling/SKILL.md). Install upstream with npx skills add Rycen7822/skill-retrieval-mcp --skill docling. Copyright stays with the author.
Docling
When to Use
Use when the user asks to convert PDF, DOCX, PPTX, HTML, Markdown, image, or scanned document content into structured Markdown, JSON, tables, or extracted text.
Do Not Use When
Do not use for repairing GB/T 7714 bibliography formatting inside a thesis DOCX or for writing a research paper.
Required Inputs
Source document path or URL, desired output format, OCR needs, table extraction requirements, and output directory.
Workflow
Inspect input document type and conversion options.
Run docling conversion with OCR/table settings when needed.
Validate extracted headings, tables, images, and metadata.
Save structured output and report conversion caveats.
Use it
Copy one of these into your project. Installing also returns the manifest and these snippets.
# after Install: the listing is in your workspace's routing pool - a plan picks it for its slotcurl-s-X POST https://api.opensmartroute.ai/api/v1/route -H 'Authorization: Bearer $OSR_API_KEY' -H 'Content-Type: application/json' -d '{"text": "...", "plan": true}'
Installed into a catalogue, chosen by a router
Install docling and it becomes one more candidate the router can pick - when it fits.
A listing is a routing target with a manifest: what it does, which domains it covers, what it costs and who publishes it. Once installed it sits beside your own models and tools, is scored like any other candidate for each request, and shows up in the trace when it wins. Ratings come from workspaces that installed it, one per account.
