Hydrogen Studies MCP server

• Registry namecom.hydrogenstudies/researchverified in the MCP registry

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

• Stateanswerschecked 2026-11-04
• Tools8
• Server calls itselfhydrogen-studies
• Server version1.0.0
• Protocol version2025-06-18

On 2026-11-04 the server answered initialize and tools/list without a sign-in. First answer in 144 ms.

• 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.

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.

• 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 answer170 ms
• Space in an agent's context≈ 2,086 tokensthe definitions of all its tools (8), which an agent carries on every turn
• Tool definitionsA · 94 of 100
• Typed results0 of 8 toolsdeclared outputSchema: results come typed, read once

• Parameters without a description: in 3 of 8

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 (25%), doing the product's job (25%), definitions (20%), context (10%), use (10%) and how readily a client connects (10%). 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 →

• search_studies

  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…

• get_study

  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…

• get_condition_evidence

  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.

• compare_evidence

  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.

• aggregate_studies

  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.

• find_related_studies

  Studies a given study cites, studies that cite it, and the most similar studies by shared health conditions.

• list_conditions

  Every health condition hub with its study count, evidence grade (where reviewed) and page URL. Use the slug as the `condition` filter.

• get_database_stats

  Size and makeup of the database: total studies, breakdown by subject and design, year range, and field coverage.