Inspect PDFs with evidence-first MCP workflows using PDF Reader MCP
Connect PDF Reader MCP when an agent needs source-backed PDF extraction with page evidence, visual crops, OCR provenance, tables, trust reports, and reviewable citations.
npx skills add agentskillexchange/skills --skill inspect-pdfs-with-evidence-first-mcp-workflows-using-pdf-reader-mcp
PDF Reader MCP exposes PDF intelligence as MCP tools for agents that need evidence, not just extracted text. Use this skill when a workflow must inspect a PDF, recover text or tables, verify answers against page regions, handle scanned pages through OCR, and attach provenance such as page numbers, bounding boxes, visual crops, and trust reports.
What this skill actually does
Invoke it instead of ordinary PDF reading when downstream decisions depend on reviewable evidence from the document. The scope boundary is MCP-based PDF inspection and provenance capture for agent workflows; it is not a generic PDF parser, OCR toolkit, or document management product listing.
Inputs and prerequisites: Node.js 22.13 or newer, npm or Docker, MCP-compatible client.
Setup notes: Run with npx using npx @sylphx/pdf-reader-mcp and add it to an MCP-compatible client, or run the Docker image with a mounted PDF workspace as documented in the upstream README.
Source and verification boundary: use https://sylphxai.github.io/pdf-reader-mcp/ as the canonical reference before running the workflow; keep commands, API calls, CLI usage, and generated outputs reviewable against that upstream source.
Framework fit: publish this as a MCP workflow only when the operator can invoke the documented toolchain directly, rather than treating the upstream project as a generic product listing.