78trust / 100

Records Vault

by eformogi.com in Security & trust

MCP serverPassing, checked 6 h ago

Consent-scoped vault access and free verification of family-issued learning records.

https://eformogi.com/api/mcp

Last 30 days

All checks passedSome failedAll failedNot checked
Uptime
100%
Response time
497 ms typical, 811 ms slowest 5%
Last check
6 h ago
Next check
in 28 min

How to call it

Add it to any MCP client that supports remote servers.

{
  "mcpServers": {
    "records-vault": {
      "type": "http",
      "url": "https://eformogi.com/api/mcp"
    }
  }
}

9 tools

  • read_profile

    Read the user's vault data for specified categories. Respects the agent policy — auto categories return data, ask categories return pending, never categories are denied.

  • list_consents

    List the user's active consent receipts — who they've shared data with and what categories.

  • create_consent

    Create a consent grant to share specific categories with an organization. Categories under "ask" policy will require user approval.

  • check_policy

    Check what data categories the agent is allowed to access and their policy level (auto/ask/never).

  • log_action

    Log an action the agent performed on behalf of the user. All MCP calls are auto-logged, but agents can add explicit action descriptions.

  • search_orgs

    Search for organizations the user can share data with.

  • verify_record

    Verify an Eformogi receipt id with no login and no API key: the same check the public /verify page performs. Returns the kind, the title or course fields, the issued date, and the signers' names and identity levels, worded exactly as the page words them. Never an email, a user id

  • how_eformogi_works

    What Eformogi is and what it costs: the thesis sentence and the pricing paragraph, verbatim, plus the key public pages.

  • read_spec

    Read the open Eformogi record standard published at /spec. With no chapter, lists every published chapter. With a chapter slug, returns that chapter's markdown text.

Security scan

  • No findings. We scan names, descriptions and tool definitions for hidden instructions and other prompt-injection patterns.

Recent checks

WhenResultHTTPTime
6 h agoPassed2001.1 s
14 h agoPassed200403 ms
23 h agoPassed200392 ms
31 h agoPassed200811 ms
40 h agoPassed200497 ms
2 days agoPassed200668 ms