Docsie MCP

- npm stdio wrapper:npx -y docsie-mcp
- PyPI / uvx stdio wrapper:uvx docsie-mcp

npm test --prefix mcp-packages/npm python -m pytest mcp-packages/pypi/tests

Live smoke test against the hosted server:

DOCSIE_MCP_ENDPOINT=https://staging.docsie.io/mcp \ DOCSIE_MCP_TOKEN=... \ node scripts/verify-live-tools.mjs

To test a different local bridge command:

DOCSIE_MCP_COMMAND="uvx docsie-mcp" \ DOCSIE_MCP_TOKEN=... \ node scripts/verify-live-tools.mjs

The live smoke verifiesinitialize,tools/list, required Docsie tools, JSON Schema shape, tool titles, and MCP tool annotations.

DOCSIE_MCP_ENDPOINT=https://staging.docsie.io/mcp \ DOCSIE_MCP_TOKEN=... \ DOCSIE_MCP_VERIFY_CALLS=1 \ DOCSIE_MCP_EXPECTED_WORKSPACE_ID=workspace_... \ node scripts/verify-live-tools.mjs

The call check verifies scoped workspace discovery throughlist_workspacesand non-mutating video credit estimation throughvideo_to_docs_estimate.

Real video-to-docs E2E smoke, for staging only:

DOCSIE_MCP_ENDPOINT=https://staging.docsie.io/mcp \ DOCSIE_MCP_TOKEN=... \ DOCSIE_MCP_E2E_CONFIRM=1 \ DOCSIE_MCP_E2E_VIDEO_URL=https://example.com/small-video.mp4 \ DOCSIE_MCP_E2E_WORKSPACE_ID=workspace_... \ node scripts/verify-video-to-docs-e2e.mjs

This creates a real video-to-docs job, polls analysis to completion, runs generation, verifies generated Docsie document links/content, and consumes Docsie AI credits. The script setsauto_generate=falseby default so the E2E controls the generation step explicitly. Optional export verification is available withDOCSIE_MCP_E2E_EXPORT_FORMATS=docx.

To resume or verify a staging run without submitting a new video:

DOCSIE_MCP_ENDPOINT=https://staging.docsie.io/mcp \ DOCSIE_MCP_TOKEN=... \ DOCSIE_MCP_E2E_CONFIRM=1 \ DOCSIE_MCP_E2E_EXISTING_FINAL_JOB_ID=job_key_... \ node scripts/verify-video-to-docs-e2e.mjs

https://app.docsie.io/.well-known/oauth-authorization-server https://app.docsie.io/.well-known/oauth-authorization-server/mcp https://app.docsie.io/.well-known/oauth-protected-resource https://app.docsie.io/.well-known/oauth-protected-resource/mcp

https://app.docsie.io/mcp/oauth/register/ https://app.docsie.io/mcp/oauth/authorize/ https://app.docsie.io/mcp/oauth/token/ https://app.docsie.io/mcp/oauth/revoke/

- Docsie user identity
- Selected organization
- Selected workspace
- Workspace membership
- Explicit permission packs

- workspace_read: read and search Docsie knowledge
- workspace_write: create and update documentation content, save comparisons, and manage persistent workspace Word fill templates
- video_docs: video-to-docs workflows, native video compare/combine jobs, Word template fill, policy checks, advanced annotations, and comparison support
- video_presentation: presentation-oriented video analysis, PPTX export workflows, and live presenter sessions
- imports: file import, website extraction, research workflows, and web-backed comparisons
- publishing: deployments, visibility, file requests, and related publishing tools
- billing: approved paid workflow options through CLI-friendly Stripe Checkout links and x402 payment requests
- agent_runs: start, monitor, cancel, and fetch remote Docsie background agent runs, with voice discovery tools when enabled for the workspace

Use the smallest permission pack set needed for your workflow.

Organization-managed service credentials are also available for shared agents that should run with a fixed Docsie account instead of requiring every downstream user to log in. Service credentials are scoped to one organization, one selected workspace, and explicit permission packs.

Some Docsie automation workflows may require additional processing capacity.

When more capacity is required, Docsie MCP can return:

- A hosted Stripe Checkout fallback
- An x402 payment request for compatible agent clients
- Structured next steps that explain the requested workflow, required authorization, and retry path

Agentic payment tools require thebillingpermission pack. Paid workflow access is granted only after payment verification and settlement.

List my Docsie workspaces. Search Docsie knowledge for onboarding API documentation. Fetch the article content for the selected Docsie article. Turn this product walkthrough video into Docsie documentation. Fill this SOP Word template from the training video. Check this onboarding video against our PII policy. Create a presentation from this product demo video. Analyze these videos and compare the generated procedures. Buy 100000 Docsie AI credits and give me the payment link. Find the customer-facing setup steps for this feature. Create a draft article from the approved workspace knowledge.

AI-powered reading triage MCP. 26 tools with a 24h burn timer. Read less, absorb more.

Docsie MCP lets you create, manage, and publish documentation, convert videos to docs, run compliance analysis, and search your knowledge base - all from your AI assistant

Connect Scrivener 3 writing projects to Claude, ChatGPT, and other AI assistants. 60+ tools for manuscript management, writing analysis, semantic search, character/plot memory, and content enhancement.

Official Notion MCP server for searching, reading, creating, and updating Notion pages, databases, and workspace content from AI agents.

Your YouTube AI Assistant To Create More Videos And Save 10+ Hours A Week

MCP server for Apple Notes with semantic search and CRUD operations. Claude searches, reads, creates, updates, and manages your Apple Notes through natural language.

Markdown-first Notion MCP server with 9 composite tools, 39 actions, and ~77% token reduction via tiered docs.

Converts video links into AI-generated Markdown notes

Video transcoding (H.264, VP9, ProRes 422, GIF), document conversion (Markdown ↔ DOCX ↔ PDF ↔ HTML ↔ XLSX), and speaker-labelled transcription for AI agents. Per-job billing, free $1 trial, no AWS or FFmpeg required.