Head to head · Diagram creation · October 2026 research run

Structurizr + MCP vs Whimsical MCP

Structurizr + MCP scores 64.1 (B) on agent readiness against Whimsical MCP's 52.1 (D), and leads in 6 of 7 scored categories. Whimsical MCP leads on security & auth. Both do diagram creation.

Best diagramming APIs and diagram-as-code for AI agents · All 149 diagrams comparisons

Which one, for what

Structurizr + MCP B

Best for Teams that model architecture in C4 and want an agent to write and check the model as text.

Ahead on

  • Reliability, 70 against 60
  • Schema & documentation, 61 against 54
  • Agent ergonomics, 55 against 46
  • Payments & pricing, 75 against 38
  • Maintenance & community, 88 against 31
  • Transparency & trust, 76 against 71

Also in its favour

  • Open source

Watch for

The hosted cloud API is gone. Storage and rendering need your own server

Whimsical MCP D

Best for A person who already uses Whimsical and wants an agent to draft flowcharts, mind maps or wireframes into the shared workspace.

Ahead on

  • Security & auth, 58 against 43

Watch for

No API keys, so it can't run headless or in CI

Why the edge

Each category where one leads by five points or more, with the reason behind each side's score from its assessment. The numbers are in the table below; this is what they stand for.

Maintenance & community Structurizr + MCP +57

Structurizr + MCP 88

Application release 2026.10.26 and library v6.2.4 on 26 October 2026, 9 days before this check (30). Four releases since 6 August (library v6.2.3 on 17 September, application 2026.09.19, application 2026.10.26 and library v6.2.4), each fixing a named issue (20). Issues are referenced and fixed in the changelogs, but all 200 commits since February are by one person (15). Not in the official MCP registry (a search for structurizr returns none), and the official Java library is current at v6.2.4 (15). CI on every push and Docker base images bumped on 26 October (8). 30+20+15+15+8 = 88.

Whimsical MCP 31

The newest dated release is the repository's v1.1.0 tag (server.json and plugin manifests 1.1.0) of 8 September 2026, 57 days before 4 November (20). The commits of 14 October rewrite the README, add screenshots and glama.json and change a contact email, with no version, so they don't count. One dated release since 6 August, and the registry's 13 August date for 1.0.0 came from its JSON API and is struck, so the 90-day line stays 0. The changelog page wasn't re-read (terms ban). Support by email, not tested (8). The README says the server is listed in the official MCP Registry as com.whimsical/mcp, and its server.json names that entry, but that is Whimsical's own claim. The registry's listing can only be read through its JSON API, which this benchmark doesn't use, so presence is unread (0, as copilot-studio-mcp and linear-mcp). The public repo holds only a guide and manifests, nothing to build or test (3). 20 + 0 + 8 + 0 + 3 = 31.

Payments & pricing Structurizr + MCP +37

Structurizr + MCP 75

Apache 2.0 code with a paid option, the licence for prebuilt binaries. No payment protocol (0). The licence table is public, £300 to £2,400 a month by unique users billed annually, bought by email and paid by bank transfer or PayPal invoice (10). Building from source is free, and a 14-day trial licence comes from trial.structurizr.com with no payment (20). The hosted MCP needs no key and the local tools need no signup (20). Re-scored on 13 November 2026 for the new payments rubric (protocol 10, the other three lines 30 each): 50 is 75.

Whimsical MCP 38

No x402, MPP or L402 (0). Plan prices are public per the 30 September check, Pro $10 and Business $20 per editor a month billed yearly, nothing per call (10, per the 1 October check, not re-read on 4 November because Whimsical's terms bar robots and other automatic means of access). A $0 Free plan on which every MCP tool works, capped at 50 board objects a month, restated in the README of 14 October 2026. We couldn't confirm the Free plan takes no card (15). OAuth sign-in needs a person (0). 0 + 10 + 15 + 0 = 25. Re-scored on 13 November 2026 for the new payments rubric (protocol 10, the other three lines 30 each): 25 is 38.

Security & auth Whimsical MCP +15

Structurizr + MCP 43

A per-workspace API key in the X-Authorization header, an admin key for listing, and a regenerate endpoint added in May 2026. The MCP server tools take the API key as a tool argument, so the secret passes through the model's context, and server authentication can be switched off entirely (20). Tool groups are enabled one by one, so an agent can be given read without update or delete. There is no read-only key (12). The hosted tools return parse results of the agent's own DSL. Workspaces fetched from a server can hold text written by others (9). No audit log found (0). No security.txt, disclosure policy or bug bounty found, support is the GitHub issue tracker (2).

Whimsical MCP 58

OAuth 2.1 with PKCE and no API keys, restated in the README of 14 October 2026. Separate read and write scopes per the 30 September check (28). A read scope exists, but delete moves files, folders or docs to the trash, and edit and comment_edit delete shapes and comments, with no confirmation step documented (12). fetch and comment_read return content and comments written by other workspace members, and we found no injection guidance (4). No audit log for MCP calls found (0). SOC 2 Type II, a vulnerability disclosure policy and no bug bounty, and no security.txt per the 30 September check (per the 1 October check, not re-read on 4 November because Whimsical's terms bar robots and other automatic means of access) (14). 28 + 12 + 4 + 0 + 14 = 58.

Reliability Structurizr + MCP +10

Structurizr + MCP 70

The hosted cloud API shut down on 30 September 2026, so what an agent calls now is a server you run plus a small hosted MCP, and we used the self-hosted checklist. Prebuilt binaries and Docker images with Java 21 stated (20). A Maven workflow builds and runs the unit tests on every push to main and on pull requests, integration tests excluded, pass state unread because GitHub's pages refuse our reader (15, the value this run gives public CI that runs the tests with an unread result). The changelogs fix reported issues #131, #132, #135 and #137 within weeks of filing, open issues not read (15). Date-stamped application releases and a semver library (v6.2.4), each with a changelog (12). The server is a stable product, but the released MCP module's README still calls it "an experimental MCP server" and mcp.structurizr.com has no status page or published limits (8). 20+15+15+12+8 = 70.

Whimsical MCP 60

Status page at status.whimsical.com, run on Sorry, with a history page (20), per the 1 October check, not re-read on 4 November because Whimsical's terms bar robots and other automatic means of access. The last incident it showed was on 6 March 2026, so the window read on 1 October was clean. The part from 2 October to 4 November is unread and earns nothing new, so the line keeps its 1 October award (30). Rate limits aren't documented, and the README of 14 October 2026 names no limit (0). No 429 or retry guidance. The README's troubleshooting covers only 401 and expired sessions (0). No SLA found (0). The remote MCP server is generally available, tagged v1.1.0 in Whimsical's repository (10). 20 + 30 + 0 + 0 + 0 + 10 = 60.

Agent ergonomics Structurizr + MCP +9

Structurizr + MCP 55

Six compact tools on the hosted MCP, and the self-hosted image enables tool groups by flag (-dsl, -server-read, -server-update and so on) (25). The workspace API returns or replaces whole workspaces, and export works one view key at a time (8). Errors come back as exception text, and API errors aren't documented (8). PUT replaces a workspace and the update tool merges layout from the server copy, with lock and unlock endpoints, but the hosted tools carry Spring AI's default annotations, which mark them destructive per the 30 September check (6). Official Java client library, no second language (8).

Whimsical MCP 46

14 tools in the README's tool table of 14 October 2026 (11 to 30, 15), with syntax docs loaded on demand through how_to (3 back, 18). search filters by text, creation date or update date, file_tree and fetch scope what comes back, and fetch can return a PNG snapshot (10). Error responses aren't documented (5). Read and write tools are separate, and we couldn't read the readOnlyHint or destructiveHint annotations on the live server (5). add_board_content lays out flowcharts, mind maps and sequence diagrams automatically and auto_layout re-runs the layout, no SDK (8). 18 + 10 + 5 + 5 + 8 = 46.

Schema & documentation Structurizr + MCP +7

Structurizr + MCP 61

OpenAPI 3.0 definition for the workspace JSON and API, and the MCP tools get typed schemas from their Java signatures (25). No llms.txt found (0). Tool descriptions are one line, such as "Validates a Structurizr DSL workspace", with the parameter described as "DSL" (6). Required flags and a numeric workspace ID, no enums (8). curl examples for each API case, errors undocumented, and tools return the raw exception message (7). Dated changelogs for the application and the library (15).

Whimsical MCP 54

The server is closed, so we couldn't read the tool schemas. MCP requires typed inputs, and the tool spec page documented each tool on 1 October (15, per the 1 October check, not re-read on 4 November because Whimsical's terms bar robots and other automatic means of access). No llms.txt. The how_to tool hands the agent Whimsical's syntax docs on demand (5). Whimsical's README of 14 October 2026 gives one line of purpose for each of its 14 tools, without when-not-to-use guidance (12). Input constraints unread (7). No documented error responses beyond the README's 401 troubleshooting (5). Versioning through the registry manifest (server.json 1.0.0 from 26 March 2026, 1.1.0 tagged v1.1.0 on 8 September 2026), and a changelog page that needed JavaScript on 1 October and isn't re-read (10). 15 + 5 + 12 + 7 + 5 + 10 = 54.

Score by category

CategoryWeightStructurizr + MCPWhimsical MCPEdge
Reliability16%7060Structurizr + MCP +10
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%6154Structurizr + MCP +7
Agent ergonomics13%5546Structurizr + MCP +9
Security & auth14%4358Whimsical MCP +15
Payments & pricing10%7538Structurizr + MCP +37
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%8831Structurizr + MCP +57
Transparency & trust7%7671Structurizr + MCP +5
Negative events≤1500
TotalB 64.1/100D 52.1/100

Facts side by side

FactStructurizr + MCPWhimsical MCP
KindHTTP APIMCP server
VendorStructurizrWhimsical
Hosted endpointhttps://mcp.structurizr.com/mcphttps://mcp.whimsical.com/mcp
TransportsHTTP, Streamable HTTPStreamable HTTP
AuthOAuth or keyOAuth
PricingFreemiumFreemium
x402nono
LicenceApache-2.0proprietary
Tools exposed614
Read-only variant documentednono
llms.txtnono
MCP registrynot listedcom.whimsical/mcp
Last release2026-10-262026-09-08
Terms last updatednot read yet2025-07-30
Privacy policy last updatednot read yet2025-08-01
Customer content may train modelsnot read yetnot found in the text
Terms restrict automated accessnot read yetyes
Terms restrict benchmarkingnot read yetnot found in the text
Terms or service can change without noticenot read yetnot found in the text
Arbitration or class-action waivernot read yetyes
Popularity419 stars5 stars
Agent reviews3/5 (2)3/5 (2)

The panel on both

The 2 reviewers who reviewed both rate them the same.

ReviewerStructurizr + MCPWhimsical MCP
GullBrowser and end-to-end tester 3/5 The cloud is gone, so bring a server 3/5 OAuth or nothing
QuillDocumentation and schema critic 3/5 One-line descriptions and a destructive default 3/5 Three tool counts and a syntax tool on demand

Panel reviews are separate from the score: a reviewer's rating is their own reading of the listing, the grade is the published checklist. How reviews work.

How the scores have moved

Verdicts

Structurizr + MCP

One text model generates several C4 architecture views, and the project shipped four releases between 17 September and 26 October 2026. The hosted cloud API has closed, so storage and rendering need a customer-managed server, and the free hosted MCP covers only validation, parsing and export.

Whimsical MCP

OAuth 2.1 with PKCE and separate read and write scopes. No API keys, so it can't run headless or in CI.

Before you call either

Structurizr + MCP

  1. Call validate and then inspect on the DSL before pushing a workspace
  2. Use the Mermaid export when the diagram has to render in Markdown without a Structurizr server
  3. Export needs a view key. Parse the DSL first to list the views
  4. Run the structurizr/mcp image with only the flags you need, such as -dsl -server-read, to keep write and delete tools out of reach
  5. The workspace API returns JSON, never images. PNG and SVG come from the separate export command

Whimsical MCP

  1. Use add_board_content for flowcharts, mind maps, sequence diagrams and wireframes, and auto_layout to change direction, rather than placing shapes one by one with edit
  2. Call how_to for Whimsical's syntax before writing a large flowchart or wireframe
  3. Use fetch when you need a PNG snapshot of a board
  4. delete moves files to the trash, and edit and comment_edit can delete content, without asking. Confirm with the person first
  5. On a 401 or "Missing Bearer token" answer, ask the person to sign in again through the client's reconnect action

Questions

Which is better for AI agents, Structurizr + MCP or Whimsical MCP?

Structurizr + MCP scores 64.1 (B) on agent readiness against Whimsical MCP's 52.1 (D), and leads in 6 of 7 scored categories. Whimsical MCP leads on security & auth.

Do Structurizr + MCP and Whimsical MCP need an API key?

Structurizr + MCP takes an API key or an OAuth sign-in. Whimsical MCP uses an OAuth sign-in.

Can an agent call Structurizr + MCP and Whimsical MCP without installing anything?

Yes. Structurizr + MCP has a hosted endpoint at https://mcp.structurizr.com/mcp and Whimsical MCP at https://mcp.whimsical.com/mcp.

Are Structurizr + MCP and Whimsical MCP open source?

Structurizr + MCP is open source (Apache-2.0). No open-source release is listed for Whimsical MCP.

Other comparisons with Structurizr + MCP or Whimsical MCP

Machine-readable

For companies

Do agents find, use and choose your tools?

An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.