Skip to content

Hydrogen Studies

by hydrogenstudies.com · remote · no sign-in to list tools · v1.1.0 · published by com.hydrogenstudies

Search, compare and summarize 2,200+ published studies on molecular hydrogen (H2) and health.

Read only

8 tools: 8 only read.

Scanned Oct 22, 2026

No check matched anything in the tool definitions. No finding is not a safety guarantee. How the checks work.

claude mcp add --transport http research https://hydrogenstudies.com/mcp

Tools

Open a tool for its reasons and full description. Descriptions are the publisher's own words, shown as they are.

8 tools

Only reads 8Claude Code: allow
  • aggregate_studiesCount matching studies grouped by year, subject, design, delivery_method, outcome, country, condition or species — for trends and 'how many' questions.No findings

    Why it is in this group

    • The description's first verb is "count": a word for reading.
    Description, as the server wrote it
    Count matching studies grouped by year, subject, design, delivery_method, outcome, country, condition or species — for trends and 'how many' questions. Unknown values are counted separately.

    Annotations the server declares

    • title: Aggregate studies
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • compare_evidenceSide-by-side evidence profiles for 2–4 groups of studies, e.g.No findings

    Why it is in this group

    • The name "compare_evidence" says "compare": a word for reading.
    Description, as the server wrote it
    Side-by-side evidence profiles for 2–4 groups of studies, e.g. inhalation vs drinking water for one condition, or two conditions. Each group is a set of filters merged over the optional shared filters.

    Annotations the server declares

    • title: Compare evidence
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • get_condition_evidenceEvidence profile for a health condition hub: how many studies, human vs animal vs in-vitro, number of human RCTs and participants, outcome mix, delivery methods, publication years,…No findings

    Why it is in this group

    • The name "get_condition_evidence" says "get": a word for reading.
    Description, as the server wrote it
    Evidence profile for a health condition hub: how many studies, human vs animal vs in-vitro, number of human RCTs and participants, outcome mix, delivery methods, publication years, data coverage, and the strongest studies.

    Annotations the server declares

    • title: Condition evidence profile
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • get_database_statsSize and makeup of the database: total studies, breakdown by subject and design, year range, and field coverage.No findings

    Why it is in this group

    • The name "get_database_stats" says "get": a word for reading.
    Description, as the server wrote it
    Size and makeup of the database: total studies, breakdown by subject and design, year range, and field coverage.

    Annotations the server declares

    • title: Database stats
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • get_studyFull record for one study: citation, design, population, H2 protocol (delivery, concentration, dosage), outcome, our plain-language summary and key finding, limitations, quality sc…No findings

    Why it is in this group

    • The name "get_study" says "get": a word for reading.
    Description, as the server wrote it
    Full record for one study: citation, design, population, H2 protocol (delivery, concentration, dosage), outcome, our plain-language summary and key finding, limitations, quality score with red flags, conditions/mechanisms, and links. Accepts an id, slug, DOI or study URL.

    Annotations the server declares

    • title: Get study
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • list_conditionsEvery health condition hub with its study count, evidence grade (where reviewed) and page URL.No findings

    Why it is in this group

    • The name "list_conditions" says "list": a word for reading.
    Description, as the server wrote it
    Every health condition hub with its study count, evidence grade (where reviewed) and page URL. Use the slug as the `condition` filter.

    Annotations the server declares

    • title: List conditions
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • search_studiesFind molecular hydrogen studies by keywords and/or filters.No findings

    Why it is in this group

    • The name "search_studies" says "search": a word for reading.
    Description, as the server wrote it
    Find molecular hydrogen studies by keywords and/or filters. Returns compact study records (design, subject, sample size, delivery method, outcome, key finding, url). Sorted by relevance when a query is given, otherwise by strength of evidence.

    Annotations the server declares

    • title: Search studies
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false

This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.

Install

Pinned to version 1.1.0. Values in angle brackets are yours to fill in.

claude mcp add --transport http research https://hydrogenstudies.com/mcp

Run in a terminal.

Claude Code permissions

Allow the reads, ask before a change, deny the deletes. Paste into .claude/settings.json.

.claude/settings.json
{
  "permissions": {
    "allow": [
      "mcp__research__search_studies",
      "mcp__research__get_study",
      "mcp__research__get_condition_evidence",
      "mcp__research__compare_evidence",
      "mcp__research__aggregate_studies",
      "mcp__research__find_related_studies",
      "mcp__research__list_conditions",
      "mcp__research__get_database_stats"
    ],
    "ask": [],
    "deny": []
  }
}

Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. It assumes the server is installed as research. Review it before you use it.