Saltar al contenido
Logo de Datronis

Servidor MCP de Datronis

Qué servidores MCP son de Datronis, del registro oficial y encontrados preguntando a su propio dominio, si responden y qué ofrecen. Un programa comprobó cada dato; nada está copiado de la página del producto.

leído el 2026-10-24

← Volver a Datronis

De quién es el servidor
  • Nombre en el registrocom.datronis/mcpverificado en el registro MCP

El nombre en el registro está bajo datronis.com. El registro solo permite usar ese espacio de nombres a quien demuestra control del dominio por DNS o HTTP.

Cuando le preguntamoscomprobado el 2026-10-24
  • Estadorespondecomprobado el 2026-10-24
  • Herramientas49
  • El servidor se llamaDatronis
  • Versión del servidor1.0.0
  • Versión del protocolo2025-06-18

El 2026-10-24 el servidor respondió a initialize y tools/list sin iniciar sesión. Primera respuesta en 537 ms.

¿Debe un agente conectarlo?Aún no está claro
  • Sus herramientas hacen find openings across several job job boards; build forms for surveys and registrations, pero no track stock reorder dates for a shop.
  • Comprobado 4 de los 7 días necesarios para juzgar su fiabilidad.

Una vez conectado

  • Para "find openings across several job job boards", la primera llamada probable es search_jobs.
  • Para "build forms for surveys and registrations", la primera llamada probable es create_form.
  • admin_create_news, admin_edit_news, create_resume, update_resume, create_memo_deck cambian datos sin iniciar sesión: cualquiera con la dirección puede llamarlas.
  • Sus definiciones ocupan unos 17337 tokens de contexto: activa solo las herramientas que necesita la tarea.
  • Parámetros sin tipo o sin descripción en 13 de 49 herramientas: revisa los argumentos antes de llamar.

Lo dicen reglas a partir de las mediciones de arriba. La primera llamada probable la elige un modelo de la lista de herramientas; lee su descripción antes de fiarte.

Medido cada díacomprobaciones diarias: 4, la primera el 2026-10-24
  • Nota INITE MCPtras una semanala nota necesita 7 comprobaciones diarias: 4 hasta ahora
  • Qué puede hacer un agente con él2 de 3 casos de uso
  • Respondiómidiendo4 de las 7 comprobaciones diarias necesarias para mostrar el porcentaje
  • Mediana de la primera respuesta733 ms
  • Espacio en el contexto del agente≈ 17.337 tokenslas definiciones de todas sus herramientas (49), que el agente lleva en cada turno
  • Definiciones de herramientasA · 90 de 100
  • Resultados tipados0 de 49 herramientasdeclaran outputSchema: el resultado llega tipado y se lee de una vez
  • Parámetros sin descripción: en 13 de 49
  • Sin indicar si solo lee o puede destruir: 15 de 49

Qué puede hacer un agente con él

  • puede · find openings across several job job boards
  • puede · build forms for surveys and registrations
  • no puede · track stock reorder dates for a shop

Cada caso de uso del producto se le preguntó a Jev junto con la lista de herramientas de este servidor: ¿puede un agente hacerlo con estas herramientas?

Conviene revisar

  • Herramientas que cambian datos, en un servidor que responde sin iniciar sesión: admin_create_news, admin_edit_news, create_resume, update_resume, create_memo_deck

Un programa comprueba el servidor una vez al día: initialize y tools/list, sin modelo. Los tokens son la longitud de las definiciones, a unos cuatro caracteres por token. La nota promedia seis comprobaciones del nombre, la descripción, el esquema y las anotaciones de cada herramienta; dice lo bien descritas que están, no si funcionan. La nota pondera solo lo medido: si responde (25%), si hace el trabajo del producto (25%), definiciones (20%), contexto (10%), uso (10%) y lo fácil que es conectarse (10%). Lo que no se midió queda fuera, y una instrucción oculta limita la nota a 40. El estado de los servidores MCP de los proveedores: método y números →

Herramientas que ofrece49 herramientas · leído el 2026-10-24
  • search_jobs

    Search job listings by keyword, location, type, or remote policy. Returns paginated results with title, company, salary, and application URL.

  • get_job

    Get full details of a job listing by its slug or ID, including description, requirements, company info, and application link.

  • get_news

    Fetch the latest news articles. Optionally filter by category, locale, or title keyword. Returns title, summary, publication date, and URL.

  • get_categories

    List all content categories available on the platform (for filtering news, jobs, etc.).

  • search_companies

    Search companies on the platform. Returns company name, industry, location, size, website, and open roles count.

  • get_platform_stats

    Get high-level statistics about the Datronis platform: total active jobs, jobs indexed in the last 7 days, companies, countries represented, news articles, and users.

  • apply_to_job

    Submit a job application on behalf of the authenticated user. Requires a personal access token (generate one at /app/api-keys/). A resume is mandatory: if no resume_id is given, the user's default (or most recently updated) resume is used…

  • list_my_resumes

    List the authenticated user's saved resumes. Requires a personal access token (generate one at /app/api-keys/).

  • get_my_applications

    Check the status of the authenticated user's job applications. Requires a personal access token (generate one at /app/api-keys/).

  • draft_news

    Draft and publish a news article. Open to anyone with a Datronis account — the personal access token is for rate-limiting and audit, not permission-gating. Always drafted in English first; an automated safety classifier must return…

  • resubmit_news

    Revise and reclassify a news article that draft_news (or a previous resubmit_news call) rejected as "flagged". Every call is its own fully logged classification attempt against the news_id's audit trail — never a silent status flip…

  • get_missing_news_translations

    List published English news articles and which locale codes they are still missing a translation for. Pass locale to focus on one language at a time. Use this to find work for submit_news_translation — this tool only reports gaps, it…

  • submit_news_translation

    Publish your own translation of an already-published English article into a new locale. Unlike draft_news's target_locales (which asks Datronis's own server to translate it), the translated title/body you pass here is published as-is…

  • admin_create_news

    Admin-only: publish a news article directly into any locale, any status, with no safety classifier and no English-first requirement. Requires an api_key belonging to an admin account — a valid but non-admin key is rejected. For non-admin…

  • admin_edit_news

    Admin-only: edit any existing news article — title, body, category, or published status — regardless of who created it or what locale it is in. Requires an api_key belonging to an admin account. Unlike resubmit_news, there is no ownership…

  • admin_translate_news

    Admin-only: durably queue server-generated translations for an existing published news article. Pass target_locales for selected languages, or omit it to fill every missing locale across all 24 supported languages. Existing translations…

  • admin_get_news_translation_status

    Admin-only: inspect translation coverage and durable queue state for a news article. Returns existing locale rows, missing locale codes, per-locale QUEUED/PROCESSING/COMPLETED/FAILED jobs, and complete=true only when all 24 languages exist.

  • create_resume

    Create a new resume for the authenticated user by directly specifying its content — name, headline, summary, contact info, work experience, education, skills, projects, certificates. Unlike the human resume builder at…

  • update_resume

    Update an existing resume's content and/or metadata (title, locale, status, default flag). By default this merges: fields you omit stay as they are, and any section array you DO send (experiences, education, skills, projects…

  • list_memo_decks

    List the authenticated user's Memo flashcard decks with card counts, how many cards are due today, and each deck's front/back language. Requires a personal access token with the "memo" scope.

  • create_memo_deck

    Create a new Memo flashcard deck, optionally with its first cards (up to 200). For language learning set front_lang/back_lang (BCP-47, e.g. "es-ES" / "en-US") so the user hears correct free text-to-speech in the browser. Requires a…

  • update_memo_deck

    Rename or restyle a Memo deck, or change its front/back language. Only fields you pass change. Requires a personal access token with the "memo" scope.

  • list_memo_cards

    List or search flashcards. With deck_id: that deck's cards (newest first). With only a query: search across all the user's decks — use this to find "the card about X" before editing or reviewing it. Every card includes `schedule`: stage…

  • get_memo_card

    Get one flashcard and exactly where it sits in its spaced-repetition cycle: stage, step (consecutive successful recalls), interval, ease, next due date, lapses, its position in the deck's study queue (queue_position of queue_size), and…

  • add_memo_cards

    Add 1–500 flashcards to a deck in one call, or update existing ones. mode "add" (default) skips exact duplicates (same front AND back, ignoring case and spacing), so re-sending a list is safe; mode "update" overwrites the answer of cards…

  • export_memo_deck

    Export every card of a deck (front, back, hint, tags) plus the deck's settings as JSON — for a backup, for converting the deck to another format for the user, or for copying it elsewhere. The same shape can be imported back with…

  • update_memo_card

    Edit a flashcard's front, back, hint or tags. Only fields you pass change; the card keeps its review schedule. Requires a personal access token with the "memo" scope.

  • get_memo_study_queue

    Get the cards due today (reviews first, then new cards) so you can quiz the user in chat: show the front, let them answer, reveal the back, then call review_memo_card with how well they did. Requires a personal access token with the…

  • review_memo_card

    Record the user's answer to a flashcard and reschedule it (SM-2 spaced repetition) — exactly the same as rating it in the web app, and it counts toward their activity calendar and streak. Also use this when the user says they already know…

  • get_memo_activity

    The user's study activity per day (reviews, correct answers, cards added), current and longest streak, and retention %. Use it to encourage them or to say which days they studied. Requires a personal access token with the "memo" scope.

  • list_forms

    List the user's Datronis forms (and forms shared with them) with status, mode, response counts and public links. Requires a personal access token with the "forms" scope.

  • get_form

    Get one form: its questions (with block ids for update_form), logic, theme, settings, validation issues and public link. Requires the "forms" scope.

  • create_form

    Create a form (as a draft). Give a plain-language description and Datronis designs the questions, or pass exact questions, or start from a template. The form is checked like any human-made form and cannot ask for passwords, card numbers…

  • update_form

    Edit a form's draft with a list of operations (add/update/remove/move questions, set title/theme/layout). All-or-nothing; returns the updated form and validation issues. Requires the "forms" scope.

  • publish_form

    Publish the current draft so respondents see it, returning the public link. If the change affects questions that already have answers, it is refused unless confirm_data_impact is true. Requires the "forms" scope.

  • set_form_status

    Close a published form (stop accepting responses) or reopen it. Requires the "forms" scope.

  • list_form_responses

    List submitted responses to a form (answers keyed by block id; get_form maps ids to questions). Reviewers only see responses assigned to them; blind review hides identities. Requires the "forms" scope.

  • get_form_response

    Get one response with answers, score, files (short-lived download links), reviews and pipeline stage. Requires the "forms" scope.

  • get_form_summary

    Aggregated results per question (counts, averages, NPS, rankings incl. instant-runoff winners) plus views, completion rate and the drop-off funnel. Requires the "forms" scope.

  • get_form_insights

    AI insights on a form's responses: a summary, main themes with verbatim quotes, mood and suspicious responses. Sensitive answers are never analysed. Requires the "forms" scope.

  • update_form_response

    Manage a response: move it to a pipeline stage (moves its Kanban card too), tag, star or add private notes. Requires the "forms" scope.

  • review_form_response

    Fill in or update your interview/application scorecard for a response. Requires the "forms" scope.

  • export_form_responses

    Export all responses as CSV text or JSON (up to 2,000 rows here; use the web app for larger exports and Excel). Requires the "forms" scope.

  • list_diagrams

    List the authenticated user's saved diagrams (name, theme, public flag, timestamps — not the full Mermaid source; call get_diagram for that). Requires a personal access token with the "diagram" scope.

  • get_diagram

    Get one diagram's full Mermaid source, theme, Global Settings, and node overrides. Requires a personal access token with the "diagram" scope.

  • create_diagram

    Create a new diagram from Mermaid source — flowcharts, sequence diagrams, mind maps, ER diagrams, Gantt charts, and anything else Mermaid supports. Requires a personal access token with the "diagram" scope.

  • update_diagram

    Update a diagram: rename it, replace its Mermaid source entirely, change its theme or Global Settings, or set its node overrides. Only fields you pass change — to edit the diagram's content, pass the complete new `code`, not a fragment…

  • delete_diagram

    Permanently delete a diagram. Cannot be undone. Requires a personal access token with the "diagram" scope.

  • duplicate_diagram

    Copy an existing diagram (including one that is public but not yours) into a new diagram you own. Requires a personal access token with the "diagram" scope.

Cómo conectar

Añade el servidor a la configuración de tu cliente MCP. Claude Desktop, Cursor y la mayoría de los demás clientes leen este formato:

{
  "mcpServers": {
    "datronis": {
      "type": "http",
      "url": "https://api.datronis.com/api/mcp"
    }
  }
}

Añadirlo

Claude Code, en una terminal

claude mcp add --transport http datronis https://api.datronis.com/api/mcp

Añadir a CursorAñadir a VS Code

claude.ai, ChatGPT y otros clientes: añade https://api.datronis.com/api/mcp como conector personalizado en los ajustes.

Cada línea es algo que un programa encontró al buscar: la entrada del registro, la respuesta del propio servidor, una especificación que se puede leer. Si falta una línea, es que no la encontramos, no que no exista.