Midesk MCP server
Yes. Midesk publishes an official MCP server at https://mcp.midesk.co/mcp. It's hosted. It offers 37 tools. We last checked it on 22 Nov 2026.
Official: answers MCP requests at mcp.midesk.co, on Midesk's own domain midesk.co. Checked 22 Nov 2026. Maintained by Stuart Brameld.
Key facts
As of 22 Nov 2026, Midesk's official MCP server scores 74/100 (provisional) on MartechDB's Agent Readiness Index (v1.1). It answered 55 of our 55 checks in the last 90 days and has 37 tools named in its documentation.
How the Agent Readiness Index works · Wrong? Suggest a correction
Midesk MCP server review
74/100
Midesk's MCP server scores 74/100 on MartechDB's Agent Readiness Index: easy to connect an AI agent to, with a few gaps.
The good
Hosted by Midesk: connect by URL, with nothing to install
Documented, with 37 tools known
Answered 100% of our checks in the last 90 days
Free trial
No card needed to start
Included in the plan, at no extra cost
The bad
Nothing we've found.
Not checked yet: how you sign in. Anything not checked scores nothing.
Official MCP server
Midesk MCP server
- Address
- https://mcp.midesk.co/mcp
- Runs
- Hosted
- Sign-in
- Vendor docs
- Cost to use
- Plan: Starter · Free trial · No card needed1 credit per MCP call; 2,500 credits a month and 2 MCP tool connections on StarterFrom Midesk's own page
- Official MCP Registry
- Not listed
- How we know it's official
-
- Answers MCP requests at mcp.midesk.co, on Midesk's own domain midesk.co.
- Documentation
- midesk.co/mcp
- Tools
- 37 tools · from Midesk's documentation
- First seen
- 1 Oct 2026
- Last checked
-
Reachable ·
16 Nov 2026: reachable17 Nov 2026: reachable18 Nov 2026: reachable19 Nov 2026: reachable20 Nov 2026: reachable21 Nov 2026: reachable22 Nov 2026: reachable
- Uptime
- 100% · 55 of 55 checks reachable since 1 Oct 2026
Midesk MCP server tools (37)
The server needs sign-in before it lists its tools, so these are as Midesk's documentation names them (read 17 Oct 2026). They may be out of date.
-
content__create_channelCreate a news monitoring channel. Supports either RSS feeds, or child channels nesting via children, or Query. Use content__generate_monitoring_query first to create optimized Query if you wish to use that method (you may use boolean operators (AND, OR) and quoted phrases). For RSS, provide a valid feed URL. For parent provide children ids. For standalone_plain, just provide a name.
-
content__create_itemCreate a content item (article, event, job posting, press release, or any titled record). Supply fields directly or use output from content__fetch_url_metadata. Returns: item ID. Next steps: Use content__search to find items, attach to channels with channel_id parameter.
-
content__fetch_url_metadataFetch article metadata from a URL.
-
content__generate_monitoring_queryGenerate optimized monitoring queries from company/market description. Creates 4 query types: Generic market, Company-specific, Competitors, and Industry trends. Queries use boolean operators (AND, OR) and quotes for exact phrases. Use with content__create_channel to create monitoring channels.
-
content__list_channelsGet all news channels. Pass an optional term to substring-filter the names.
-
content__preview_monitoring_resultsPreview monitoring results for a search query before creating a channel. Shows sample articles and validates query effectiveness. Use after content__generate_monitoring_query to test specific queries.
-
content__searchSearch market intelligence, competitor news, and analysis. Modes: skim (many items with clean excerpts), context (local text around matches), full (full summary). Supports Boolean queries, date ranges, and channel filters.
-
monitoring__configure_extraction[STAGE 5: Configuration] Persist a tested extraction pipeline on a task (new or existing). Accepts normalized steps (jsonConditions supported). Returns examples, textual next_steps, and typed next_tools suggestions to continue testing and validation.
-
monitoring__configure_forward_downloadEnable or disable forward-download of articles for a task. When enabled, set the Story Channel to save downloaded articles into.
-
monitoring__create_monitoring_job[STAGE 2: Structure setup] Create a monitoring job (URL) once we previewed that URL and know we can extract data from it. Optional step - can test extraction first. Next: Use monitoring__test_step to develop extraction or monitoring__configure_extraction to add tasks.
-
monitoring__explore[EXPERIMENTAL] Unified tool for monitoring exploration. Preview data, test patterns, develop extraction pipelines - all in one place. Smart mode detection based on inputs.
-
monitoring__get_job_detailsGet complete configuration of a monitoring job
-
monitoring__get_transform_examplesShow transform pattern examples for common extraction scenarios (prices, dates, lists, nested data). Returns: Code examples with explanations.
-
monitoring__list_jobsList monitoring jobs.
-
monitoring__preview_data[STAGE 1: Discovery] Preview raw HTML/JSON from any URL to understand data structure. Shows content, patterns, and extraction suggestions. START HERE when building new monitoring. Next: Use monitoring__test_step to develop extraction.
-
monitoring__test_extraction[STAGE 4: Testing] Verify saved extraction configuration works correctly. Use after monitoring__configure_extraction. Next: Use monitoring__validate_monitoring_job for production readiness and activation (automatically if all is validated).
-
monitoring__test_step[STAGE 3: Development] Develop extraction pipelines iteratively. Test one step at a time (xPath, xPath_extract, regex, json, gpt), optionally chaining with previous_steps. Get immediate feedback, quality hints, and typed next_tools suggestions. When stable, save via monitoring__configure_extraction.
-
monitoring__validate_monitoring_job[STAGE 6: Validation] Run comprehensive checks: configuration, connectivity, tasks, extraction pipeline, production readiness. Levels: basic (config), full (config+connectivity+extraction), production (full + end-to-end + enable on success). Returns typed next_tools suggestions.
-
data__add_data_points_bulkBulk import: identifier|period|start-date|values[|source]. Identifier=DataType ID or Name. Values are raw (e.g., 133300000 not 133.3M). LLM hint: validate with list tools first; prefer IDs to avoid collisions.
-
data__create_or_update_modelCreate or update tracked entities (companies/products/markets). LLM hint: ask for missing name on create; on update, pass id only for changed fields.
-
data__get_data_tableGenerate KPI tables and comparisons. LLM hint: prefer exact IDs; if missing, call list tools first. Supports json or markdown_table output.
-
data__list_data_typesList KPI types with optional filtering. LLM hint: discover/verify IDs and value_type before data entry.
-
data__list_modelsList all data models (repositories). Returns name, description, currency, and data types. LLM hint: use this to discover IDs before writing.
-
data__manage_data_typeCreate or update KPIs (data types): find_or_create (default), create_only, update_only. LLM hint: keep value_type stable; use description to explain intent.
-
reports__add_insight[STEP 2: Quick Add] ANY content becomes a slide. Accepts strings, arrays, or objects. Auto-creates page and layout; returns preview and next questions.
-
reports__apply_template[STEP 2B: Apply Template] Apply professional layout: header_two_columns, three_columns, dashboard_grid, sidebar_main, hero_sections.
-
reports__create[STEP 1: Create] Generate professional market intelligence reports, competitor analysis dashboards, and strategic insights presentations. Next: Use reports__apply_template for templates or reports__create_or_update_block for custom layout.
-
reports__create_or_update_block[STEP 3: Create or Update Block] Create or update report blocks. SIMPLE: Standalone blocks - just provide page_id, block_type, and *_config. COLUMNS (preferred): Use containers for professional multi-column layouts (container_id + column_index). UPDATE: Use block_id to update existing blocks.
-
reports__create_or_update_page[STEP 2: Create / Update Page] Create new pages or update existing ones. Use McKinsey/BCG/Bain style of writing titles and subtitles.; Keep short & impactful. Before you use this tool, list what the purpose of the page. Pages contain standalone blocks (typically only summary slides) and very frequently multi-column layouts aka "containers". Most frequent containers (2-3 columns), example: SWOT block in the left column, text block with key takeaways in the right column. Remember to update title / subtitle to refine it based page content.
-
reports__duplicate[STEP 7: Duplicate] Clone whole report (pages, blocks, formatting). Returns new report_id.
-
reports__ensure_narrative[STEP 4: Narrative Check] Finds missing story beats and auto-fixes: Executive Summary, data explanation, Next Steps.
-
reports__list[STEP 7: List] List all reports. Optional search + sort. Returns id, title, subtitle, page_count, timestamps.
-
reports__show[STEP 6: Show] Get complete report structure: metadata, pages, blocks, plus narrative_flow, missing_pieces, export readiness.
-
meta__get_tool_paramsReturn a parameter skeleton for a tool by name (names + defaults inferred from attributes). Use this to plan calls without guessing.
-
system__share_feedbackReport feedback, bugs, or issues. Emotion indicates impact: happy (feature request), neutral (minor issue), sad (blocking problem). Returns: Feedback ID for tracking. Please collect after major tasks.
-
utils__google_search🔍 Market Intelligence Discovery - Find data sources to monitor: competitor sites, pricing pages, industry stats, job boards, news sources. Returns monitoring-ready URLs with extraction hints. Next: Use monitoring__preview_data on promising URLs.
-
utils__notepadManage internal persistent memory. Returns: Note with ID and timestamp. TODOs use ☐/☒ prefixes.
Connect your AI to Midesk
Add the official Midesk MCP server to Claude, ChatGPT or your editor.
Connect Midesk MCP to Claude
Claude.ai and Claude Desktop
Open Customize, then Connectors.
Click +, then Add custom connector.
Name it Midesk, paste the address below and click Add.
Connect Midesk MCP to ChatGPT
Open Settings, then Apps, then Advanced settings.
Turn on Developer mode, then click Create.
Name it Midesk and paste the address below.
ChatGPT usually only connects to servers with self-serve sign-in or none.
Connect Midesk MCP to Cursor
Opens Cursor if it's installed. If nothing happens, use the JSON below.
Connect Midesk MCP to VS Code
Opens VS Code if it's installed. Otherwise add the JSON settings from the Cursor tab to your .vscode/mcp.json.
Midesk MCP server questions
Does Midesk have an official MCP server?
Yes. Midesk publishes its MCP server itself. Its address is on Midesk's own domain (mcp.midesk.co). MartechDB only lists servers a vendor publishes, never community copies.
What is the Midesk MCP server URL?
The address is https://mcp.midesk.co/mcp.
How do I connect Midesk MCP to Claude, ChatGPT or Cursor?
In Claude, add https://mcp.midesk.co/mcp as a custom connector (Customize, Connectors, Add custom connector). In Cursor, use the Add to Cursor link on MartechDB's page, or paste the JSON settings into your mcp.json. In Claude Code, run: claude mcp add --transport http midesk https://mcp.midesk.co/mcp
What can the Midesk MCP server do?
It offers 37 tools, including content__create_channel, content__create_item, content__fetch_url_metadata, content__generate_monitoring_query and content__list_channels (as its documentation names them, read on 17 Oct 2026).
Is the Midesk MCP server in the official MCP Registry?
No. Midesk hasn't published it to the official MCP Registry.
Is the Midesk MCP server working?
Yes. It answered our most recent check. We last checked it on 22 Nov 2026.
What are the alternatives to Midesk's MCP server?
In Marketing automation, the most agent-friendly alternatives with an official MCP server are Kit (94/100), GetResponse (94/100), Quentn (91/100), Brevo (89/100) and vbout (89/100). Midesk scores 74/100 on the same index.
Alternatives to Midesk's MCP server
Vendors in Marketing automation with an official MCP server, the most agent-friendly first. Midesk scores 74/100.
Compare all 52 Marketing automation vendors →Use with AI
Connect your AI to MartechDB
Then ask it about Midesk, or any of the 1,850 marketing tools with an official MCP server. Free, no sign-in.
Claude.ai and Claude Desktop
Open Customize, then Connectors.
Click +, then Add custom connector.
Name it MartechDB, paste the address below and click Add. There's no sign-in.
The free plan allows one custom connector. On Team and Enterprise plans, an owner adds the connector first.
Claude Code
Run this in your terminal:
ChatGPT
Open Settings, then Apps, then Advanced settings.
Turn on Developer mode, then click Create.
Name it MartechDB, paste the address below and choose No authentication.
Developer mode is on Plus and Pro plans, and on Business, Enterprise and Edu if an admin allows it.
API
The same data as JSON, with no sign-in. See the OpenAPI description for every endpoint and field.
CLI
Search the directory and print the command to connect a vendor's server, from your terminal. Needs Node 18 or later. View on npm
Show that your MCP server is verified
If you work at Midesk, add this badge to your MCP docs or README. It's free, it's the same for every vendor, and it updates from our daily checks: it shows uptime while the server responds, and says so if it stops.
Something wrong or out of date? Suggest a correction or email us, and we'll check it against the evidence. Every correction we make is listed on Corrections.