Skip to content
DigitalPublic logo

DigitalPublic MCP server

Which servers in the official MCP registry are DigitalPublic's own, whether they answer, and what they offer. A program checked each of these, nobody copied them from a product page.

read 2026-10-24

← Back to DigitalPublic

Whose server it is
  • Registry namecom.digitalpublic/commercialverified in the MCP registry

The registry name sits under digitalpublic.com. The registry only lets a publisher use that namespace after proving control of the domain through DNS or HTTP.

When we asked itchecked 2026-10-24
  • Stateanswerschecked 2026-10-24
  • Tools27
  • Server calls itselfdigitalpublic-commercial
  • Server version2026-09-10-trial-active-projects-v1
  • Protocol version2025-06-18

On 2026-10-24 the server answered initialize and tools/list without a sign-in. First answer in 545 ms.

Should an agent connect itNot proven either way yet
  • Whether its tools do the product's job has not been judged yet.
  • Checked on 2 of the 7 days it takes to judge reliability.

Once connected

  • Parameters are untyped or undescribed in 22 of 27 tools: check arguments before calling.

Said by rules from the measurements above. The likely first call is a model's pick from the tool list; read its description before relying on it.

Measured dailydaily checks: 2, the first on 2026-10-24
  • INITE MCP scoreafter a weeka score needs 7 daily checks: 2 so far
  • Answeredmeasuring2 of the 7 daily checks it takes to print a share
  • Median first answer620 ms
  • Space in an agent's context≈ 3,040 tokensthe definitions of all its tools (27), which an agent carries on every turn
  • Tool definitionsB · 86 of 100
  • Typed results27 of 27 toolsdeclared outputSchema: results come typed, read once
  • Parameters without a description: in 22 of 27

Worth a look

None of the five warning signs in these definitions.

A program checks the server once a day: initialize and tools/list, no model. Tokens are the definitions' length at about four characters a token. The grade averages six checks of every tool's name, description, schema and annotations; it says how well the tools are described, not whether they work. The score weighs what was measured and nothing else: answering (30%), doing the product's job (25%), definitions (20%), context (10%) and use (15%). A part that was not measured is left out, and a hidden instruction caps the score at 40. The state of vendors' MCP servers: method and numbers →

Tools it offers27 tools · read 2026-10-24
  • dp_commercial_get_company_profile

    Return the verified legal identity, corporate contacts, and evidence URLs for DigitalPublic.

  • dp_commercial_ask_navigator

    Match a prospective need with public DigitalPublic capabilities, illustrative use cases and plans. It has no access to customer projects or private data.

  • dp_commercial_get_capability_matrix

    List read, preparation, supervised-write, approval, plan, and module capabilities with explicit qualifications.

  • dp_commercial_get_data_processing_profile

    Explain controller and processor roles, public AI training policy, private customer data exclusions, controls, DPA, privacy, and subprocessor evidence URLs.

  • dp_commercial_compare_integration_options

    Provide an objective comparison of direct provider MCP servers and DigitalPublic across setup, governance, cross-channel work, and tradeoffs.

  • dp_commercial_estimate_roi

    Calculate only the value of entered time savings, plan cost, net value, and break-even hours. It never forecasts sales or advertising performance.

  • dp_commercial_list_use_cases

    List clearly labeled illustrative use cases without presenting them as customer results or performance claims.

  • dp_commercial_get_service_status

    Return the current component health and timestamp. Status is measured by an independent monitor hosted outside the platform. Components with no recent evidence are reported as 'unknown' — never assume 'unknown' means healthy. Does not…

  • dp_commercial_get_uptime_history

    Return historical uptime for a component, always alongside its data coverage. A high uptime with low coverage is not meaningful, so both values must be reported together. Returns 'insufficient_data' when coverage is too low to publish a…

  • dp_commercial_start_live_trial

    Start an idempotent 14-day trial with up to three active real projects, 2,000 total units, up to five connected tools and supervised write access. Sensitive actions still require human approval. A single passwordless email verifies…

  • dp_commercial_get_live_trial_status

    Resume a trial using its limited status token. Returns stable state and nextAction, remaining days, units and connection capacity, plus only safe human links. It never returns email, provider tokens, OAuth URLs, or provider payloads.

  • dp_commercial_get_install_options

    Return publication-aware installation options for the authenticated DigitalPublic operational product in OpenAI, Claude Code, and generic MCP clients. Directory URLs remain null until publication is explicitly enabled.

  • dp_commercial_get_onboarding_guide

    Return the goal-oriented sequence an agent should follow, including every mandatory human checkpoint and the handoff from DigitalPublic Commercial to DigitalPublic.

  • dp_commercial_describe_platform

    Explain what DigitalPublic does for humans and AI agents.

  • dp_commercial_list_offers

    List current plans, EUR and USD prices, limits, seats, modules, and checkout mode.

  • dp_commercial_get_offer

    Return one current plan by id.

  • dp_commercial_compare_offers

    Compare two or more plans using the canonical catalog.

  • dp_commercial_assess_fit

    Recommend a plan from project count, team size, live-data need, distributor status, and write requirement.

  • dp_commercial_list_integrations

    List integrations included in commercial V1. Use trialChannels, not the connector id, as channels for dp_commercial_start_live_trial.

  • dp_commercial_get_connection_requirements

    Explain authentication modes, scopes, human prerequisites, review state and known limitations for one DigitalPublic connector.

  • dp_commercial_get_security_profile

    Explain project isolation, scopes, approvals, token revocation, and audit controls.

  • dp_commercial_get_checkout_readiness

    Report whether the real trial, self-service checkout, and Enterprise proposal onboarding are currently available.

  • dp_commercial_get_sample_report

    Return an illustrative synthetic report sample without customer data. This is not a service trial and must never be represented as measured customer performance.

  • dp_commercial_start_checkout

    Compatibility checkout for paid plans and Enterprise proposals. Use dp_commercial_start_live_trial for evaluation. A human must verify email and complete any payment.

  • dp_commercial_get_checkout_status

    Return the non-sensitive status of a historical checkout or proposal request. For the real trial use dp_commercial_get_live_trial_status.

  • dp_commercial_claim_credentials

    Legacy credential claim for still-valid historical paid checkout intents. Live trials never deliver an operational credential in chat; install DigitalPublic and use project OAuth instead.

  • dp_commercial_request_proposal

    Request an email-verified Enterprise proposal. This does not create a contract or charge.

How to connect

Add the server to your MCP client config. Claude Desktop, Cursor and most other clients read this shape:

{
  "mcpServers": {
    "commercial": {
      "type": "http",
      "url": "https://platform.digitalpublic.com/api/mcp-commercial"
    }
  }
}

Add it

Claude Code, in a terminal

claude mcp add --transport http digitalpublic https://platform.digitalpublic.com/api/mcp-commercial

Add to CursorAdd to VS Code

claude.ai, ChatGPT and other clients: add https://platform.digitalpublic.com/api/mcp-commercial as a custom connector in the settings.

Other servers listed as DigitalPublic's
  • com.digitalpublic/marketing-operationsanswers, requires sign-inverified in the MCP registry

OpenAPI spec

20 paths, read 2026-10-24. Open the spec ↗

Each line is something a program found when it went looking: the registry entry, the server's own answer, a spec that parses. A missing line means we did not find it, not that it does not exist.