Skip to main content
Glama
davidmosiah

Google Health MCP

by davidmosiah

Google Health MCP Capabilities

google_health_capabilities
Read-onlyIdempotent

Explain supported Google Health data, privacy boundaries, beta status, and recommended agent workflow.

Instructions

Explain supported Google Health data, privacy boundaries, beta status and recommended agent workflow.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
response_formatNomarkdown

Output Schema

TableJSON Schema
NameRequiredDescriptionDefault
projectYes
mcp_nameYes
creatorYes
unofficialYes
statusYes
beta_noticeYes
api_boundaryYes
auth_modelYes
privacy_modesYes
supported_dataYes
recommended_agent_flowYes
client_aliasesYes
contribution_pathsYes
linksYes

Implementation Reference

  • Tool handler registration for google_health_capabilities. The handler calls buildCapabilities() and formats the response via makeResponse with a bullet list showing project, status, api_boundary, recommended_first_tools, and docs link.
    server.registerTool("google_health_capabilities", {
      title: "Google Health MCP Capabilities",
      description: "Explain supported Google Health data, privacy boundaries, beta status and recommended agent workflow.",
      inputSchema: ResponseOnlyInputSchema.shape,
      outputSchema: CapabilitiesOutputSchema.shape,
      annotations: { readOnlyHint: true, destructiveHint: false, idempotentHint: true, openWorldHint: false }
    }, async ({ response_format }) => {
      const capabilities = buildCapabilities();
      return makeResponse(capabilities, response_format, bulletList("Google Health MCP Capabilities", {
        project: capabilities.project,
        status: capabilities.status,
        api_boundary: capabilities.api_boundary.source,
        recommended_first_tools: "google_health_connection_status, google_health_data_inventory, google_health_daily_summary",
        docs: capabilities.links.docs
      }));
    });
  • Input schema (ResponseOnlyInputSchema) for the tool — only accepts an optional response_format parameter.
    export const ResponseOnlyInputSchema = z.object({
      response_format: ResponseFormatSchema
    }).strict();
  • Output schema (CapabilitiesOutputSchema) defining the shape of the capabilities response: project, mcp_name, creator, api_boundary, auth_model, privacy_modes, supported_data, recommended_agent_flow, client_aliases, contribution_paths, and links.
    export const CapabilitiesOutputSchema = z.object({
      project: z.string(),
      mcp_name: z.string(),
      creator: z.object({ name: z.string(), github: z.string() }).strict(),
      unofficial: z.boolean(),
      status: z.string(),
      beta_notice: z.string(),
      api_boundary: z.object({ source: z.string(), raw_definition: z.string(), does_not_include: z.array(z.string()) }).strict(),
      auth_model: z.object({ type: z.string(), token_storage: z.string(), recommended_redirect_uri: z.string(), default_scopes: z.array(z.string()) }).strict(),
      privacy_modes: z.array(z.object({ mode: PrivacyModeValueSchema, use_when: z.string() }).strict()),
      supported_data: z.array(z.object({ name: z.string(), examples: z.array(z.string()), tools: z.array(z.string()) }).strict()),
      recommended_agent_flow: z.array(z.string()),
      client_aliases: z.record(z.string(), z.unknown()),
      contribution_paths: z.array(z.string()),
      links: z.record(z.string(), z.string())
    }).passthrough();
  • buildCapabilities() function that returns the full static capabilities object containing project metadata, API boundary info, auth model, privacy modes, supported data types, recommended agent flow, client aliases, contribution paths, and reference links.
    export function buildCapabilities() {
      return {
        project: "google-health-mcp-unofficial",
        mcp_name: "io.github.davidmosiah/google-health-mcp",
        creator: { name: "David Mosiah", github: "https://github.com/davidmosiah" },
        unofficial: true,
        status: "beta",
        beta_notice: GOOGLE_HEALTH_BETA_NOTICE,
        api_boundary: {
          source: "Official Google Health API v4 with Google OAuth 2.0",
          raw_definition: "Raw means the full JSON response returned by supported Google Health API v4 endpoints under https://health.googleapis.com.",
          does_not_include: [
            "Google Fit REST API legacy endpoints",
            "Android-only Health Connect on-device storage",
            "raw accelerometer/device telemetry",
            "private Google endpoints",
            "write/upload actions by default",
            "medical diagnosis or treatment guidance"
          ]
        },
        auth_model: {
          type: "Google OAuth 2.0 authorization code with offline refresh tokens",
          token_storage: "Local token file with user-only permissions",
          recommended_redirect_uri: "http://127.0.0.1:3000/callback",
          default_scopes: DEFAULT_SCOPES
        },
        privacy_modes: [
          { mode: "summary", use_when: "Default-safe interpretation with identifiers and source details minimized." },
          { mode: "structured", use_when: "Normalized Google Health data points and rollups for agents." },
          { mode: "raw", use_when: "The user explicitly needs upstream Google Health payloads for debugging or deep analysis." }
        ],
        supported_data: [
          { name: "Identity, profile and settings", examples: ["Google Health user id", "legacy Fitbit id presence", "profile", "units", "timezone"], tools: ["google_health_get_identity", "google_health_get_profile", "google_health_get_settings"] },
          { name: "Data point queries", examples: ["steps", "sleep", "heart-rate", "weight", "exercise"], tools: ["google_health_list_data_points", "google_health_reconcile_data_points"] },
          { name: "Daily rollups", examples: ["steps", "distance", "total-calories", "active-zone-minutes", "weight", "daily-resting-heart-rate"], tools: ["google_health_daily_rollup"] },
          { name: "Physical-time rollups", examples: ["hourly steps", "heart-rate windows", "distance windows"], tools: ["google_health_rollup"] },
          { name: "Agent summaries", examples: ["daily summary", "weekly review", "wellness context"], tools: ["google_health_daily_summary", "google_health_weekly_summary", "google_health_wellness_context"] }
        ],
        recommended_agent_flow: [
          "Call google_health_agent_manifest when installing or operating inside a server agent such as Hermes.",
          "Call google_health_connection_status before calling Google Health data tools.",
          "If setup is incomplete, guide the user through Google Cloud setup, OAuth auth and doctor.",
          "Use google_health_data_inventory to pick data types and understand endpoint/filter naming.",
          "Use google_health_daily_summary or google_health_weekly_summary before low-level endpoint tools.",
          "Treat health data as sensitive; avoid raw payloads unless explicitly requested.",
          "Use Google Health as trend context, not medical diagnosis. Escalate symptoms or abnormal vitals to clinicians."
        ],
        client_aliases: {
          hermes: {
            tool_prefix: "mcp_google_health_",
            direct_tools: [
              "mcp_google_health_google_health_agent_manifest",
              "mcp_google_health_google_health_connection_status",
              "mcp_google_health_google_health_data_inventory",
              "mcp_google_health_google_health_daily_summary",
              "mcp_google_health_google_health_weekly_summary"
            ],
            reload_command: "/reload-mcp",
            gateway_restart_required_for_data_access: false
          }
        },
        contribution_paths: [
          "Add real-account fixture coverage as Google Health v4 stabilizes.",
          "Add source-family-specific UX for Pixel Watch, Fitbit and Google first-party sources.",
          "Add webhook/subscriber support after read-only flows are proven.",
          "Add optional write tools only behind explicit opt-in and safety gates."
        ],
        links: {
          github: "https://github.com/davidmosiah/google-health-mcp",
          docs: "https://wellness.delx.ai/connectors/google-health",
          npm: "https://www.npmjs.com/package/google-health-mcp-unofficial",
          google_health_docs: "https://developers.google.com/health",
          google_health_reference: "https://developers.google.com/health/reference/rest",
          google_health_scopes: "https://developers.google.com/health/scopes",
          google_health_data_types: "https://developers.google.com/health/data-types",
          google_cloud_console: "https://console.cloud.google.com/apis/library/health.googleapis.com"
        }
      };
  • Tool name listed in the STANDARD_TOOLS array within the agent manifest, confirming it is part of the standard tool set.
    const STANDARD_TOOLS = [
      "google_health_agent_manifest",
      "google_health_cache_status",
      "google_health_capabilities",
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare a safe, read-only, idempotent tool. The description adds valuable context about the content (data, privacy, beta status, workflow) beyond what annotations provide. No contradictions.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Single sentence, front-loaded with key purpose, no wasted words.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given low complexity (1 optional param, output schema present), the description covers all key aspects: what data, privacy boundaries, beta status, and workflow guidance. Output schema handles return value documentation.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters1/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

The only parameter (response_format) has no schema description (0% coverage) and is not mentioned in the tool description. The agent gets no guidance on how to use the parameter or its valid values.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description uses specific verbs ('Explain') and specifies the resource ('supported Google Health data, privacy boundaries, beta status and recommended agent workflow'), clearly distinguishing it from siblings like data_inventory or privacy_audit.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description implies the tool is for initial orientation and suggests a 'recommended agent workflow', but does not explicitly state when to use it vs. alternatives or when not to use it.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Install Server

Other Tools

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/davidmosiah/google-health-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server