Mithrandir — free MCP gateway test

Observe MCP tools, verify repeated reads, then try bounded exact reuse for free.

Should I use this

Quality & Safety

B
Description quality
100%
Schema completeness
53%
Naming quality
80%
Poisoning risk
100%
Permission match
100%
Protocol compliance
100%

Based on automated analysis of tool definitions and protocol compliance.

Context Cost

~542Tokens (tool definitions)
~1.1 KBTypical response size
Minimal attention impact (0.42% of 128k context)

This is the approximate number of tokens consumed each time the server's tools are loaded into a model's context. Higher counts reduce the attention available for other tasks.

Install

One-Click Install

Add this to your `claude_desktop_config.json` file:

{
  "mcpServers": {
    "mithrandir": {
      "url": "https://mithrandir-production.up.railway.app/mcp"
    }
  }
}

Remote endpoints

https://mithrandir-production.up.railway.app/mcpstreamable-http

What it can do

Tool inventory

Tools (3)

🟢 Read-only🟡 Write🔴 Delete⚪ Unknown
🟢mithrandir_demo(scenario)

Run a free isolated synthetic Reuse Autopilot demonstration. No production account, upload, payment or private data is used. Compare unchanged, one-partition-changed and missing-source behavior before onboarding.

Input Schema

{
  "type": "object",
  "properties": {
    "scenario": {
      "enum": [
        "unchanged",
        "one_partition_changed",
        "missing_source"
      ],
      "default": "one_partition_changed"
    }
  },
  "required": [],
  "additionalProperties": false
}
🔴mithrandir_gateway_trial_start(name, upstream_url, policy_version, accept_terms, cache_ttl_seconds, ...)

Start a bounded free gateway test. Connect your public HTTPS MCP server, then approve safe reads and verify repetition before free Optimize activation.

Input Schema

{
  "type": "object",
  "properties": {
    "name": {
      "type": "string",
      "pattern": "^[A-Za-z0-9_.-]{1,80}$"
    },
    "upstream_url": {
      "type": "string",
      "minLength": 12,
      "maxLength": 1000
    },
    "policy_version": {
      "type": "string",
      "minLength": 1,
      "maxLength": 80
    },
    "accept_terms": {
      "const": true
    },
    "cache_ttl_seconds": {
      "type": "integer",
      "minimum": 30,
      "maximum": 86400,
      "default": 300
    },
    "safe_read_tools": {
      "type": "array",
      "items": {
        "type": "string",
        "pattern": "^[A-Za-z0-9_.:/-]{1,160}$"
      },
      "maxItems": 100,
      "default": []
    },
    "tool_costs_microusd": {
      "type": "object",
      "additionalProperties": {
        "type": "integer",
        "minimum": 0,
        "maximum": 1000000000000
      },
      "default": {}
    },
    "min_shadow_matches": {
      "type": "integer",
      "minimum": 1,
      "maximum": 20,
      "default": 2
    },
    "idempotency_key": {
      "type": "string",
      "minLength": 32,
      "maxLength": 128
    }
  },
  "required": [
    "name",
    "upstream_url",
    "policy_version",
    "accept_terms",
    "idempotency_key"
  ],
  "additionalProperties": false
}
🟢mithrandir_security_profile

Free machine-readable description of the v17 Cognitive Membrane trust model: origin-bound authority, task-scoped exact-call capabilities, replay protection, egress control and two-phase effectful actions. This is a security contract description, not proof that a deployment is vulnerability-free.

Input Schema

{
  "type": "object",
  "properties": {},
  "required": [],
  "additionalProperties": false
}

Community

Rate this Server

Evidence

Recent observations

verifiedversion not recorded3 tools