Head to head · Content management · October 2026 research run

dotCMS vs Payload

dotCMS scores 57.7 (C) on agent readiness against Payload's 50.1 (D), and leads in 2 of 7 scored categories. Payload leads on reliability, schema & documentation, payments & pricing and transparency & trust. Both do content management.

Best headless CMS and website publishing for AI agents · All 136 cms comparisons

Which one, for what

dotCMS C

Best for Enterprises already on dotCMS who want an agent to manage content and pages.

Ahead on

  • Security & auth, 63 against 57

Also in its favour

  • Runs on your own machine
  • No incidents deducted, where Payload loses 15 points for them

Watch for

Pre-1.0 package, with the dev docs still pointing at the beta tag

Payload D

Best for Teams that build on Next.js and want the content model in TypeScript, with drafts, versions and localisation in the free core.

Ahead on

  • Reliability, 76 against 47
  • Schema & documentation, 70 against 65
  • Payments & pricing, 45 against 35
  • Transparency & trust, 65 against 54

Also in its favour

  • Open source

Watch for

38 security advisories on OSV in the 12 months to 4 November 2026, 7 critical, 28 published on 6 and 7 October

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.

Reliability Payload +29

dotCMS 47

Read with the local-software lines, since the MCP server is a stdio package the owner runs against their own instance. Official npm package with Node 20 or later and Bun stated (20). The monorepo has CI and the MCP app has vitest tests and smoke tests; whether they pass on main wasn't read (12 of 25). Open issues weren't read because the GitHub API refused us for its rate limit (10 of 25). 0.1.0 and 0.1.1 without release notes for the package (5 of 15). Pre-1.0, and the dev docs still call it beta (0). 20+12+10+5+0 = 47.

Payload 76

Read with the local-software lines, since the graded surface is software its owner hosts. Official npm packages, with Node.js ^22.23.4 || >=24.22.1 in engines and in the installation docs since 3.92.0 (20). A public ci workflow runs unit, type, integration and end-to-end tests on every push to 3.x. GitHub's pages and API couldn't be read on 4 November, so the result is unread (15 of 25). 499 issues were open on 8 October, 140 of them awaiting triage, with 225 opened and 116 closed since 10 July 2026, not re-read (12 of 25). A written semver policy marks breaking commits with ! and lists them in release notes, but lets security fixes and experimental APIs change in minor releases (14 of 15). 3.x is the stable line, with 4.0 in beta (15).

Transparency & trust Payload +11

dotCMS 54

dotCMS core is under the Business Source License 1.1, source-available rather than OSI, with an additional use grant under $5 million; the MCP package is GPL-3.0 on npm (20 of 30). The privacy policy of 8 September 2026 covers website visitors only; no cloud data-handling terms were found (10 of 30). A deprecations page, an LTS lifecycle and dated changelogs (16 of 20). Telemetry and sub-processors weren't established (5 of 20). 20+10+16+5 = 51.

Payload 65

MIT for the core and official packages, with LICENSE.md in the repository (30). A self-hosted install keeps content on its owner's servers. The privacy policy dates from 28 March 2024, states no retention periods that we could find and doesn't mention Figma. The only published terms cover Payload Cloud, which is closed to new projects (15 of 30). A written versioning policy defines breaking changes and how they're announced, with a migration guide per major. No end-of-support date for 3.x was found (13 of 20). Telemetry is on by default, documented as completely anonymous and turned off with telemetry: false. The docs don't list the fields sent. The 3.92.1 source sends versions, adapters, plugins, locales, a random install ID and project, domain and user IDs hashed with the install's own secret (16 of 20).

Payments & pricing Payload +10

dotCMS 35

Scored with the self-hosted rule, with prices from the cloud plans. No x402, MPP or L402 (0). Cloud prices are on request only (0). Free with every capability under $5 million in revenue under the BSL, with no card (20). Self-hosting needs no dotCMS account, and tokens can be minted by API on one's own instance (15 of 20). 0+0+20+15 = 35.

Payload 45

Scored with the self-hosted rule. No x402, MPP or L402 (0). The software is free under MIT, which the get-started page states. The only paid option, Enterprise, is priced by sales, and Payload Cloud is closed to new projects, so 15 of 20. No card and no account at Payload are needed (20). npx create-payload-app installs it, but the docs create REST and MCP API keys in the admin panel, so half (10).

Security & auth dotCMS +6

dotCMS 63

JWT API tokens per user with role permissions and an expiry, created only by administrators (25 of 30). A read-only token can be made through a restricted role, but the MCP server asks for write permission; workflows support multiple-approval steps (14 of 20). Content returned from the CMS reaches the agent, and the sandbox isolates the token, with no prompt-injection guidance (5 of 15). Audit logging for API calls wasn't established in the pages read (5 of 15). A valid security.txt with PGP, a responsible disclosure policy with bounties, LTS security backports and a public list of known security issues, which is drawn by script and couldn't be read. No certification was found (14 of 20). 25+14+5+5+14 = 63.

Payload 57

REST keys are per user, revocable and regenerable, shown once, and limited by that user's access control down to field level. MCP keys carry per-collection, per-operation switches. Both travel only in the Authorization header. Neither expires, and there is no OAuth (24 of 30). A find-only MCP key is a config switch away and the collection's access control still runs under it. Nothing asks a person to approve a delete or a publish in the free edition (14 of 20). Documents hold text written by other people, and no prompt-injection guidance was found. overrideResponse can strip fields before they reach a model (3 of 15). Versions record each change, and onEvent and verboseLogs expose MCP calls to the owner's own logging. No built-in audit log (7 of 15). SECURITY.md gives a reporting address and, since 16 October 2026, says Payload pays no monetary rewards and runs no bug bounty. Advisories are published on GitHub with affected versions, patches and workarounds. No security.txt or SOC 2 statement was found (9 of 20).

Schema & documentation Payload +5

dotCMS 65

The search tool exposes the instance's own OpenAPI spec, so the contract is machine-readable, though the tools themselves take JavaScript code (20 of 25). llms.txt and Markdown pages on dev.dotcms.com (10). The README says what each tool is for and gives examples; when not to use them isn't stated (12 of 20). Code strings rather than typed parameters for search and execute (4 of 15). Examples for both tools; error behaviour of the sandbox isn't documented (9 of 15). dotCMS changelogs are dated; the MCP package has no changelog (10 of 15). 20+10+12+4+9+10 = 65.

Payload 70

No OpenAPI file is published or generated by the core packages, because endpoints depend on each project's config. Every MCP tool carries a typed input schema built from the collection's fields, and the GraphQL schema is generated from the same config (20 of 25). llms.txt per major version, llms-full.txt and a Markdown copy of every docs page (10). Default MCP tool descriptions are one line, such as Create a document in a collection., unless the owner writes one in the plugin config. The docs tell owners to do so (10 of 20). Field inputs are typed with bounds on limit and depth, but where is JSON passed as a string. select became an object in the MCP find tools in 3.91.0 (10 of 15). REST operations show a request and a response each. No reference page for error responses or status codes was found (8 of 15). Semver, release notes on GitHub and separate docs per major version. REST paths carry no version (12 of 15).

Score by category

CategoryWeight this rundotCMSPayloadEdge
Reliability16%204776Payload +29
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%16.26570Payload +5
Agent ergonomics13%16.26864dotCMS +4
Security & auth14%17.56357dotCMS +6
Payments & pricing10%12.53545Payload +10
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%8.87578Payload +3
Transparency & trust7%8.85465Payload +11
Negative events≤150-15
TotalC 57.7/100D 50.1/100

Facts side by side

FactdotCMSPayload
KindMCP serverHTTP API
VendordotCMS LLCPayload CMS, Inc. (Figma)
Hosted endpointno (local only)no (local only)
TransportsstdioHTTP, Streamable HTTP
AuthAPI keyAPI key
PricingFreemiumFree
x402nono
LicenceBUSL-1.1MIT for the core and the official packages. Enterprise add-ons are sold separately through sales
Tools exposed7none
Read-only variant documentednono
llms.txtyesyes
Last release2026-09-022026-10-27
Terms last updatedno date givenno document linked
Privacy policy last updated2026-09-082024-03-28
Customer content may train modelsnot found in the textnot read yet
Terms restrict automated accessnot found in the textnot read yet
Terms restrict benchmarkingnot found in the textnot read yet
Terms or service can change without noticenot found in the textnot read yet
Arbitration or class-action waivernot found in the textnot read yet
Popularity43 npm/wk45k stars, 1.1M npm/wk

How the scores have moved

Verdicts

dotCMS

dotCMS's MCP server keeps the tool list to seven by letting the agent search the instance's OpenAPI spec and run API calls in a sandbox that never sees the token. It is a pre-1.0 package that needs a write token, and cloud prices are on request.

Payload

Payload generates REST, GraphQL and MCP interfaces from one typed config, with drafts, restorable versions and per-key MCP permissions in the free MIT core. The security record is the limit. 38 advisories were published in the 12 months to November 2026, most on 6 and 7 October, so an install older than 3.90.0 is exposed. No OpenAPI file is published.

Before you call either

dotCMS

  1. Search the spec with search before writing an execute call, and return only the fields you need
  2. Use a token for a dedicated user and role, not an administrator's
  3. Confirm with the user before execute calls that publish, archive or delete content
  4. Keep results small with pick and first, since output is capped at about 25,000 characters
  5. Use download_assets for theme folders so files land on disk instead of in context

Payload

  1. Check the installed version first. Anything below 3.90.0 carries published critical advisories, and 3.92.0 and later need Node.js 22.23.4 or 24.22.1, so ask the owner to upgrade before writing
  2. Send REST keys as Authorization: {collection-slug} API-Key {key} and MCP keys as Authorization: Bearer {key}. The two key kinds are separate
  3. To publish, set _status: 'published' in the data. The draft parameter only relaxes validation and chooses where an update is written
  4. Upload files with multipart POST to the upload collection, with other fields as JSON in _payload. No MCP upload tool is documented
  5. Roll back with POST /api/{collection-slug}/versions/:id after listing versions. Versions exist only where the collection config enables them

Questions

Which is better for AI agents, dotCMS or Payload?

dotCMS scores 57.7 (C) on agent readiness against Payload's 50.1 (D), and leads in 2 of 7 scored categories. Payload leads on reliability, schema & documentation, payments & pricing and transparency & trust.

Do dotCMS and Payload need an API key?

Both need an API key.

Can an agent call dotCMS and Payload without installing anything?

dotCMS runs on your own machine, with no hosted endpoint listed. No hosted endpoint is listed for Payload.

Are dotCMS and Payload open source?

No open-source release is listed for dotCMS. Payload is open source (MIT for the core and the official packages. Enterprise add-ons are sold separately through sales).

Other comparisons with dotCMS or Payload

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.