Skip to content
Patient Recruitment for Clinical Research logo

Patient Recruitment for Clinical Research MCP server

Which MCP servers are Patient Recruitment for Clinical Research's own, from the official registry and from asking its own domain, 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 Patient Recruitment for Clinical Research

Whose server it is
  • Registry nameai.hipa/clinical-trialsverified in the MCP registry

The registry name sits under hipa.ai. 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
  • Tools8
  • Server calls itselfhipa-trials
  • Server version0.1.2
  • Protocol version2025-06-18

On 2026-10-24 the server answered initialize and tools/list without a sign-in. First answer in 290 ms.

Should an agent connect itConnect it, with care
  • Its tools do screen trial applicants against protocol eligibility criteria, but not track applicant status and notes across team members; publish a site page for each recruiting study.

Once connected

  • For "screen trial applicants against protocol eligibility criteria", the likely first call is clinicaltrials_find_eligible_trials.
  • Parameters are untyped or undescribed in 2 of 8 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: 10, the first on 2026-10-24
  • INITE MCP score80 of 100answers 100 · does the job 33 · definitions 96 · context 93 · connects 100
  • What an agent can do with it1 of 3 use cases
  • Answered100% of days10 of 10 daily checks
  • Median first answer264 ms
  • Space in an agent's context≈ 3,845 tokensthe definitions of all its tools (8), which an agent carries on every turn
  • Tool definitionsA · 96 of 100
  • Typed results8 of 8 toolsdeclared outputSchema: results come typed, read once
  • Parameters without a description: in 2 of 8

What an agent can do with it

  • can do · screen trial applicants against protocol eligibility criteria
  • cannot · track applicant status and notes across team members
  • cannot · publish a site page for each recruiting study

Each of the product's use cases, put to Jev against this server's tool list: can an agent do it with these tools?

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 (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 →

Tools it offers8 tools · read 2026-10-24
  • clinicaltrials_count_trials

    Return only the count of active US clinical trials matching the same filter set accepted by clinicaltrials_search_trials. Use this when the caller needs a population estimate (e.g. 'how many trials are recruiting in California for type 2…

  • clinicaltrials_find_eligible_trials

    Match patient demographics (age, sex, conditions, location) to currently recruiting US clinical trials. Conditions are curated slugs, labels or trial-record condition names (discover them via clinicaltrials_list_conditions). Maps ageYears…

  • clinicaltrials_get_data_freshness

    Returns the timestamp at which the underlying ClinicalTrials.gov dataset was last refreshed on the server. Use this to disclose the data age to end users or to decide whether to re-query for fresher results.

  • clinicaltrials_get_trial

    Fetch the full clinical trial record by NCT ID. Returns title, status, sponsor, summary, conditions, interventions, arms, design, eligibility (age, sex, healthy volunteers and the full inclusion/exclusion criteria text), primary outcomes…

  • clinicaltrials_list_conditions

    List the curated conditions the trial search tools filter on, with the slug to pass as `condition` (clinicaltrials_search_trials, clinicaltrials_count_trials) or in `conditions` (clinicaltrials_find_eligible_trials). Pass `query` to find…

  • clinicaltrials_list_top_conditions

    List the most common conditions across active US clinical trials in a US state or city, with trial counts. Use this when the caller wants to see what is well-represented in a region before searching. Each entry carries the curated `slug`…

  • clinicaltrials_list_top_facilities

    List named research facilities running active US clinical trials in a city, ranked by trial count. Returns each facility's name, ZIP, the top conditions they study, and one example NCT ID for quick verification. Use this when the caller…

  • clinicaltrials_search_trials

    Search active US clinical trials by location, condition slug, phase, study type, intervention type, recruitment status, sex eligibility, age bucket, healthy-volunteer flag, and facility name. Returns up to 50 trials per page with an…

How to connect

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

{
  "mcpServers": {
    "clinical-trials": {
      "type": "http",
      "url": "https://hipa.ai/api/mcp/mcp"
    }
  }
}

Add it

Claude Code, in a terminal

claude mcp add --transport http hipa https://hipa.ai/api/mcp/mcp

Add to CursorAdd to VS Code

claude.ai, ChatGPT and other clients: add https://hipa.ai/api/mcp/mcp as a custom connector in the settings.

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.