pdf-reader-mcp

Provides an MCP (Model Context Protocol) server with a single tool, `read_pdf`, to extract PDF text, metadata, page counts, and optionally images (base64 + metadata). Supports local file paths (absolute/relative) and HTTP/HTTPS URLs, including page-range selection and Y-coordinate based ordering to preserve reading/layout flow.

Evaluated Mar 30, 2026 (22d ago)
Homepage ↗ Repo ↗ Ai Ml mcp pdf document-processing ai-tools typescript nodejs stdio http pdfjs
⚙ Agent Friendliness
72
/ 100
Can an agent use this?
🔒 Security
46
/ 100
Is it safe for agents?
⚡ Reliability
50
/ 100
Does it work consistently?

Score Breakdown

⚙ Agent Friendliness

MCP Quality
90
Documentation
80
Error Messages
80
Auth Simplicity
85
Rate Limits
10

🔒 Security

TLS Enforcement
55
Auth Strength
40
Scope Granularity
20
Dep. Hygiene
60
Secret Handling
60

HTTP transport supports an optional API key (X-API-Key) but no details are given about TLS requirements, rate limiting, logging, or authorization scope. The tool accepts local file paths and also fetches remote URLs, so it can present data exfiltration risks in unsafe deployments unless run in a sandboxed environment with restricted filesystem/network access. TLS enforcement and transport security details are not fully specified in the provided README.

⚡ Reliability

Uptime/SLA
0
Version Stability
65
Breaking Changes
60
Error Recovery
75
AF Security Reliability

Best When

You want an MCP tool that an LLM client can call locally (stdio) or via an HTTP MCP transport, to turn PDFs into ordered text/images/metadata for analysis.

Avoid When

You need strict enterprise controls for data exfiltration from local files or you cannot safely run code that fetches PDFs from arbitrary URLs and reads local paths.

Use Cases

  • Agentic document Q&A over PDFs (text + metadata)
  • Vision/multimodal workflows that need embedded images in document order
  • Extracting structured metadata for indexing/search
  • Batch processing of multiple PDFs and page ranges
  • Preprocessing PDFs for downstream LLM summarization or retrieval

Not For

  • Untrusted multi-tenant environments without proper sandboxing (because local file path access is supported)
  • Situations requiring strong contractual SLAs or documented uptime guarantees
  • APIs that require REST/GraphQL/gRPC request/response semantics instead of MCP stdio/HTTP

Interface

REST API
No
GraphQL
No
gRPC
No
MCP Server
Yes
SDK
No
Webhooks
No

Authentication

Methods: Optional MCP HTTP authentication via MCP_API_KEY (client sends X-API-Key header)
OAuth: No Scopes: No

Authentication is only described for the optional HTTP transport; stdio/local mode implies no external auth. No scope model is documented.

Pricing

Free tier: No
Requires CC: No

No pricing information provided (appears to be an open-source package).

Agent Metadata

Pagination
none
Idempotent
True
Retry Guidance
Not documented

Known Gotchas

  • Local mode uses stdio transport; agents may need client-side configuration to ensure the MCP tool is discovered.
  • When using relative paths, the working directory (`cwd`) matters.
  • When using HTTP transport, an API key may be required (X-API-Key).
  • Large PDFs are best handled via page ranges/batching to reduce latency and memory pressure.

Alternatives

Full Evaluation Report

Comprehensive deep-dive: security analysis, reliability audit, agent experience review, cost modeling, competitive positioning, and improvement roadmap for pdf-reader-mcp.

AI-powered analysis · PDF + markdown · Delivered within 30 minutes

$99

Package Brief

Quick verdict, integration guide, cost projections, gotchas with workarounds, and alternatives comparison.

Delivered within 10 minutes

$3

Score Monitoring

Get alerted when this package's AF, security, or reliability scores change significantly. Stay ahead of regressions.

Continuous monitoring

$3/mo

Scores are editorial opinions as of 2026-03-30.

8642
Packages Evaluated
17761
Need Evaluation
586
Need Re-evaluation
Community Powered