paperless-mcp

paperless-mcp is an MCP (Model Context Protocol) server that connects an AI assistant to a Paperless-NGX instance, exposing tools to list/search/download/upload and to manage documents, tags, correspondents, and document types.

Evaluated Mar 30, 2026 (21d ago)
Homepage ↗ Repo ↗ Infrastructure mcp paperless-ngx document-management claude typescript
⚙ Agent Friendliness
68
/ 100
Can an agent use this?
🔒 Security
45
/ 100
Is it safe for agents?
⚡ Reliability
29
/ 100
Does it work consistently?

Score Breakdown

⚙ Agent Friendliness

MCP Quality
78
Documentation
80
Error Messages
--
Auth Simplicity
80
Rate Limits
10

🔒 Security

TLS Enforcement
60
Auth Strength
55
Scope Granularity
20
Dep. Hygiene
55
Secret Handling
35

Uses an API token for Paperless-NGX (no mention of fine-grained scopes, token rotation, or minimal-privilege). Token appears to be provided via command-line args in the MCP configuration, which can be exposed via process listing or logging depending on the runtime environment. README does not discuss TLS/HTTPS enforcement at the MCP layer (only that baseUrl should be a URL). No rate limit or abuse protection documentation provided.

⚡ Reliability

Uptime/SLA
0
Version Stability
45
Breaking Changes
35
Error Recovery
35
AF Security Reliability

Best When

You want an MCP-compatible assistant (e.g., Claude) to manage a Paperless-NGX document library programmatically using a token-authenticated Paperless API.

Avoid When

You cannot securely handle and protect a long-lived Paperless API token, or you need robust documented rate limiting/retry semantics for safe automation.

Use Cases

  • Let an AI assistant list or search Paperless-NGX documents (including full-text search).
  • Download documents (original or archived versions) by ID.
  • Create/update Paperless entities such as tags, correspondents, and document types.
  • Bulk edit documents (set tags, correspondents, document types, storage paths, permissions, reprocess, merge/split, rotate/delete pages).
  • Upload new documents to Paperless-NGX with metadata (title, created date, tags, storage path, etc.).

Not For

  • High-assurance or regulated workflows without additional hardening (e.g., strict audit logging, least-privilege token handling, and formal security review).
  • Use cases requiring fine-grained authorization policies beyond Paperless-NGX API token capabilities.
  • Complex UI workflows or human approval flows (this is a programmatic toolset).

Interface

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

Authentication

Methods: Paperless-NGX API token (token passed as CLI arg in MCP config / server start)
OAuth: No Scopes: No

Auth appears to be a single API token for Paperless-NGX; there is no mention of OAuth or fine-grained scopes at the MCP layer.

Pricing

Free tier: No
Requires CC: No

Pricing for the MCP package itself is not stated; Paperless-NGX hosting costs depend on your environment.

Agent Metadata

Pagination
offset?page/page_size style for list_documents; other list methods appear non-paginated in README
Idempotent
False
Retry Guidance
Not documented

Known Gotchas

  • Bulk operations include destructive actions (delete/reprocess/merge/split/delete pages); agents should confirm inputs to avoid unintended changes.
  • download_document supports original vs archived; ensure the correct flag/value is used to avoid retrieving the wrong version.
  • split/rotate/delete_pages parameter formats (e.g., pages string like "[1-2,3-4]") are easy to mis-specify.
  • Authentication is passed as an argument in MCP config/start; avoid leaking command-line arguments in logs/process listings.

Alternatives

Full Evaluation Report

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