Skip to main content
Notra runs a hosted MCP server at https://mcp.usenotra.com/mcp (streamable HTTP) so AI agents can work with your workspace. Connected agents can manage brand identities, generate, edit, schedule and publish content, preview and publish posts on Notra Sites, follow their builds and manage their authors, connect integrations and manage schedules, event triggers, webhooks and skills. They can also use every GEO feature: projects, prompts, competitors, scans, visibility, content briefs, agent readiness, AI traffic and website analytics. This page covers two things:
  • Notra’s MCP server. Give an AI client (Claude, Cursor, ChatGPT, Codex and others) access to Notra.
  • External MCP servers. Give Notra’s chat agent access to your own tools by connecting servers under Integrations > MCP Servers.

Authentication

The server accepts bearer credentials in the Authorization header, either as an OAuth access token or as a Notra API key.

Connect an AI client

With OAuth

Add the server URL without any headers, and the client detects the OAuth metadata and opens the sign-in flow.
In Claude and ChatGPT, the quickest way to connect is the Claude connector or the ChatGPT plugin. You can also add https://mcp.usenotra.com/mcp as a custom connector in either app and complete the sign-in when prompted.

With an API key

If your client supports add-mcp, install the server with a header override:
Or add the header to the client config by hand:

Local stdio server

The same server is published on npm as @usenotra/mcp for clients that only speak stdio. It needs Node.js 20 or newer and an API key in NOTRA_API_KEY:
The hosted server gets updates first and the npm package can lag behind it, so prefer the hosted URL when your client supports remote servers.

Load only the tools you need

The server exposes every tool by default. Many clients load all tool definitions into the agent’s context up front, so if you only use part of Notra, pick a toolset to keep that context small: Workspace tools and submit_feedback are always available. Add the toolsets query parameter to the hosted URL, or set NOTRA_MCP_TOOLSETS for the npm package. Separate several toolsets with commas.
If you already added notra to Claude Code, run claude mcp remove notra first, because claude mcp add does not replace an existing server with the same name. In Cursor, edit the existing entry. The server rejects an unknown toolset name with an error instead of silently hiding tools. OAuth sign-in works the same with the query parameter.

Available tools

Tool names match exactly what the server registers, although some clients prefix them with the server name. Workspace
  • whoami
  • list_workspaces (only on connections that can reach several workspaces)
Posts
  • list_posts
  • get_post
  • create_post
  • update_post
  • delete_post
  • generate_post
  • get_post_generation_status
  • create_post_pull_request
  • get_post_schedule
  • schedule_post
  • cancel_post_schedule
When update_post changes the title, slug or markdown of a blog post or changelog with a linked GitHub pull request, Notra saves the article and updates that PR using the same publishing logic as the dashboard, without creating a new PR. Status-only changes do not trigger a GitHub update. If the PR update fails, the tool returns an error stating that Notra saved the post. A rate-limited GitHub sync returns 429 with Retry-After, so wait before retrying the content update. A timeout returns 504 with an unknown sync outcome, and because the PR update may still finish, check the PR before retrying, as you should after any other unconfirmed sync error. If the PR is closed or merged, resolve it in the dashboard before Notra can sync it again. To turn a pull request into a post, call generate_post with pullRequest and, if you like, a brief in your own words. Poll get_post_generation_status until it reports a postId, then call create_post_pull_request to open a draft pull request with the post. Calling it again updates that pull request. Sites
  • list_sites
  • get_site
  • list_site_deployments
  • get_site_deployment
  • deploy_site
  • redeploy_site_deployment
  • preview_post_on_site
  • get_site_post_preview
  • publish_post_to_site
publish_post_to_site always opens or updates a pull request, even when the repository publishes directly from the dashboard, so the post goes live only after someone merges it. get_site_deployment returns compiler diagnostics and the newest build log output. Domains, settings, rollbacks and deletion stay in the dashboard. Sites authors
  • list_site_authors
  • upsert_site_author
  • delete_site_author
  • import_site_authors
  • set_site_author_page
  • publish_site_design
Author changes are saved to the site’s blog.json as unpublished Design changes, the same ones the dashboard’s Design tab shows, and go live when publish_site_design or the dashboard publishes them. publish_site_design refuses while other files on the site have unpublished changes, so work you staged in the dashboard is never published by accident. To credit a site author on a blog post that publishes to the site, pass its site:<id> ref in authorIds of update_post. The Sites tools need the sites.read and sites.write scopes. Brand identities
  • list_brand_identities
  • get_brand_identity
  • update_brand_identity
  • delete_brand_identity
  • generate_brand_identity
  • get_brand_identity_generation_status
  • list_brand_colors
  • create_brand_color
  • update_brand_color
  • delete_brand_color
Integrations
  • list_integrations
  • list_github_activity
  • create_github_integration
  • delete_integration
Schedules and event triggers
  • list_schedules
  • create_schedule
  • update_schedule
  • delete_schedule
  • run_schedule
  • list_schedule_runs
  • list_event_triggers
  • get_event_trigger
  • create_event_trigger
  • update_event_trigger
  • delete_event_trigger
Webhooks
  • list_webhook_endpoints
  • delete_webhook_endpoint
  • list_webhook_deliveries
  • get_webhook_delivery
  • retry_webhook_delivery
Create webhook endpoints in the dashboard, because each endpoint gets a signing secret. Skills
  • list_skills
  • get_skill
  • create_skill
  • update_skill
  • delete_skill
GEO projects and settings
  • list_projects
  • get_project
  • create_project
  • update_project
  • delete_project
  • get_geo_settings
  • list_geo_engines
  • update_geo_settings
GEO diagnostics
  • get_geo_snapshot
  • get_geo_changes
  • get_geo_prompt_history
  • get_geo_sentiment
  • get_geo_sentiment_analysis
  • list_geo_sentiment_evidence
  • list_geo_shelf_sources
Start with get_geo_snapshot. It returns a compact diagnosis of a project and the recommended next actions. GEO prompts and sequences
  • list_geo_prompts
  • create_geo_prompt
  • update_geo_prompt
  • delete_geo_prompt
  • import_geo_prompts
  • list_geo_prompt_suggestions
  • accept_geo_prompt_suggestion
  • dismiss_geo_prompt_suggestion
  • list_geo_sequences
  • create_geo_sequence
  • update_geo_sequence
  • delete_geo_sequence
  • run_geo_sequence
GEO competitors
  • list_geo_competitors
  • upsert_geo_competitor
  • suggest_geo_competitors
  • delete_geo_competitor
  • import_geo_competitors
GEO scans and visibility
  • create_geo_scan
  • rescan_geo_prompt
  • list_geo_scans
  • get_geo_scan
  • get_geo_visibility_overview
  • get_geo_visibility_timeseries
  • list_geo_prompt_result_summaries
  • get_geo_prompt_result_detail
  • get_geo_competitor_share
  • get_geo_language_share
  • get_geo_competitor_detail
GEO content gaps and briefs
  • list_geo_content_gaps
  • update_geo_content_gap
  • list_geo_content_briefs
  • plan_geo_content_brief
  • get_geo_content_brief
  • update_geo_content_brief
  • approve_geo_content_brief
Agent readiness
  • get_geo_agent_readiness
  • start_geo_agent_readiness_scan
AI traffic
  • get_geo_traffic_overview
  • get_geo_traffic_log
  • list_geo_traffic_journeys
  • get_geo_traffic_journey
  • list_geo_traffic_pages
  • get_geo_traffic_journey_stats
  • get_geo_ingest_setup (includes the public notra.js web key and pinned script URLs)
Issue and rotate ingest tokens and the web key in the dashboard. Website analytics
  • get_geo_web_analytics
  • get_geo_web_analytics_breakdowns
  • list_geo_analytics_imports
get_geo_web_analytics returns views, visitors, sessions and AI-referred visitors against the previous period, and get_geo_web_analytics_breakdowns returns pages, sources, countries, devices, languages, outbound links, Core Web Vitals and JavaScript errors. History imports need provider credentials, so start them in the dashboard; the MCP server only reports their progress. MCP server analytics
  • get_geo_mcp_analytics summarizes how agents use your own MCP server: tool calls, errors, latency, clients, models and tools nobody calls.
  • get_geo_mcp_tool_analytics shows one tool’s failures and which tools agents call right before and after it.
These read the events your server sends through @usenotra/sdk/mcp. Captured arguments, responses and end-user IDs stay in the dashboard. Feedback
  • list_feedback, get_feedback and update_feedback work with the feedback your own users send you.
  • submit_feedback sends a bug report, feature request, question or praise to the Notra team and needs no credentials. See Agent feedback.
create_geo_scan, rescan_geo_prompt, run_geo_sequence and plan_geo_content_brief use billed AI credits and can take minutes, so the server instructs agents to confirm with you before starting them.

Connect external MCP servers to Notra

Notra’s chat agent can also use tools from MCP servers you run or subscribe to. Open Integrations > MCP Servers in the dashboard (or press C on that page) and add a server.
1

Describe the server

Give it a name, the server URL (for example mcp.example.com/mcp) and a short description of what tools or context Notra should use it for. The agent uses the description to decide when to use the server.
2

Choose authentication

Pick one of three options:
  • None for public servers without credentials.
  • API key to send one or more custom headers, such as Authorization: Bearer ....
  • OAuth to sign in with the server’s own OAuth flow, which Notra opens in a popup.
3

Test and save

Notra tests the connection before saving and reports whether it could reach the server. With OAuth, the button reads Connect & Authorize and Notra saves the server once authorization completes.
After saving, Notra indexes the server’s tools and shows the tool count on the server card, where you can refresh the tool index, enable or disable the server, reauthorize an expired OAuth grant or delete it. The Notra chat agent searches the indexed tools and activates the ones it needs for a conversation.

Composio

If you want to use the Notra MCP server through Composio, upvote the request at request.composio.dev/boards/tool-requests/posts/notra so the Composio team sees demand for a Notra integration.

Notes

  • Prefer OAuth for personal clients so tokens rotate and can be revoked per session.
  • Keep API keys in a server-side or local secret store, and do not commit MCP config files that contain live bearer tokens.
  • Revoke and replace a key immediately if it is exposed.
Last modified on November 15, 2026